Free spectral emission (FSE) of RhB molecules were collected from RhB + PVA films coated on glass substrates using the SPCE setup as mentioned above without prism coupling. On exciting the molecules at 532 nm, FSE were observed at wavelengths above 550 nm. A broad spectrum spanning in the range of 550 nm to 750 nm was recorded and is presented in Fig. 1(c). The same dye molecules in the vicinity of metal thin films show different pattern as compared to its FSE. To observe this variation, RhB + PVA films were coated on top SPR metal sensors Ag, Au and Cu and their respective SPCE spectra were recorded and analysed. SPCE spectra for silver based SPCE sensor with RhB + PVA film thickness of 220 nm (Ag220) was first studied in detail. The spectra were collected at angles from 200 to 800 with an interval of 10. It was observed that the emission was only for a few angles between 400 and 450 and no emission was observed for other angles. It shows a highly anisotropic emission as compared to its free spectral emission which was isotropic in nature.
All the collected SPCE spectra were organised as a matrix with wavelength along x-direction, angle along y-direction and emission intensity as colour scale along z-direction. The same is presented as an image plot in Fig. 2 (a). A controlled emission within 400 and 450 was evident in the image plot. SPR reflection curves were generated by transfer matrix model simulations25 to correlate the SPCE emission with the SPR reflection. On comparing the emission spectra of Ag220 with its simulated SPR reflection image (Fig. 2. (b)), it is evident that the emission steering matches well with the p-polarised dispersion curve between 400 and 450. The SPR image (Fig. 2(b)) also shows s-polarised dispersion curve between angles 480 and 580 which is missing in SPCE image (Fig. 2(a)). This might be due to the weak coupling of emission with the corresponding s-polarised SPR.
SPCE emission spectra for the angles between 420 and 470 is presented in Fig. 2(c). On analysing individual spectra obtained in between 420 and 470 we could see that the peak wavelengths differ from its FSE. Peak wavelength of FSE of RhB was at 600 nm and for the SPCE spectra at 420 it was at 650 nm (Fig. 2(c)). As the angle was increased from 420 to 430 the peak wavelength shifted to 625 nm. The peak shit decreased as angle increased from 420 to 470. Once the peak shift reached 600 nm then no further shift was observed. On comparing SPCE with FSE emission intensity, more than an order of magnitude in emission intensity was evident. In contrast to emission peak intensity enhancement the full width half maximum (FWHM) of SPCE emission was observed to be reduced as compared to FSE emission. The observed emission tuning and steering in SPCE matches exactly with the simulated SPR curves and hence a perfect correlation between SPCE and SPR could be confirmed. The emission tuning confirms a strong coupling of dye molecules with the plasmons within the metal dielectric interface. Figure 2(d) show polar plot of the SPCE emission at different angles for a particular wavelength (610 nm) (corresponding to peak FSE wavelength) which confirms the SPCE emission directionality. Thus, from our studies, the plasmon coupled emission shows emission tuning apart from the typical emission steering effects reported previously18.
Further studies on SPCE tuning were carried out to examine the effect of the emission tuning on the thickness of the dielectric layer. Figure 3(a) shows the experimental SPCE image plot of Ag600 (t = 600 nm) in RK configuration and Fig. 3(b) shows the corresponding simulated SPR image.
The results of the sample Ag600 are shown in Fig. 3. For dielectric layer thickness less than 200nm we could observe only p-polarised emission. As the thickness was increased to 600 nm both s-and p-polarised emission were dominant (Fig. 3(a)). Thus, dual polarised SPCE could be realised for thicker dielectric layers. As thickness of the dielectric layer increases multimode SPCE were observed for both s and p polarised emissions (Fig. 3(b)). The number of SPCE emission modes exactly matches with the simulated SPR reflection images as shown in Fig. 2(a) and Fig. 2(b) respectively. It is to be noted that as compared to Ag220 where we could observe only p-polarised mode, in Ag600 we could observe that both s- and p-polarised modes were prominent with considerable emission intensity. All the modes show SPCE tuning apart from emission enhancement and emission steering. S1 and S2 modes tune between 420 and 440 and between 580 and 620 respectively. P1 and P2 modes get tuned between 460 and 520, and between 640 and 700 respectively. Corresponding to the angular tuning, emission peak wavelength also gets tuned as follows. The first plasmonic mode S1, shows SPCE tuning from 570 nm to 670 nm for an angular span of 20, the second mode P1, shows SPCE tuning from 560 nm to 680 nm for an angular span of 60. Similarly, the S2 mode shows a tuning from 590 nm to 750 nm for 40 shift and P2 mode shows a tuning from 580 nm to 830 nm for an angular span of 60. In addition to the observed emission steering and tuning, SPCE enhancement was also prominent in Ag600. These observed results confirm that the surface plasmon coupling gets improved on increasing dielectric thickness of the sensor chip. The enhancement in SPCE intensity reveals that the p-polarisation modes have better plasmon coupling than the s-polarisation modes. These observations again support the direct correlation between SPCE and SPR and the effect of SPCE on the dielectric thickness coated on the SPR sensor chips. Figure 3(d) presents the polar plot of Ag600 at 610 nm wavelength, which confirms dual polarised emission apart from the emission steering, tuning and enhancement as compared Ag220. The increase in sharpness of these lines with thickness is due to strong plasmon coupling. In our previous work we could observe only p-polarised mode SPCE for all the dielectric layer thicknesses (210 nm to 690 nm). In our present studies we could observe dual polarised SPCE emission and emission tuning. We have detected two modes for both S and P polarisations when the sample thickness was increased from 220nm to 600nm. The observed emission results had a direct correlation with the simulated SPR image (Fig. 3(b)).
SPCE investigation studies were extended on gold and copper SPR metal sensors. Dielectric layer thickness of 70, 270 and 590 nm on Au were coated for the SPCE studies (named as Au70, Au270 and Au590). The results of Au70 are shown in Fig. 4. The experimental image plot of the sample Au70 and its simulated SPR image are shown in Fig. 4(a) and Fig. 4(b) respectively. The SPCE from Au70 sample was observed to be highly directional and it shows tuning effect (SPCET). Results show only p-polarised SPCE and the emission peak was observed to be tuned from 780 nm to 560 nm as the angle of detection was varied from 500 to 750. As the samples thickness was very thin, plasmon coupling effects were feeble and hence the SPCE studies showed low emission intensities. The Fig. 4(c) represents plasmon coupled emission curves extracted from the image plot (Fig. 4(a)) between angles 500 and 750 and Fig. 4(d) represents the angle scanned SPCE of the sample as a polar plot.
Thicker dielectric layers on Au metal films were also investigated to compare the higher order SPCE modes. The image plot and its simulated SPR image of Au270 are shown in Fig. 5(a) and Fig. 5(b) respectively. Similar to Ag metal sensors, Au also showed dual polarised surface plasmon coupled emissions. On comparing simulated SPR image (Fig. 5(b)) with the SPCE image (Fig. 5(a)), it was evident that the two SPCE emissions in the angular regions 420 to 480 and 560 to 660 corresponds to p and s polarised emissions respectively. Tuning was observed for both s- and p-polarised emissions. For s-polarised emission the peak emission got tuned from 580 nm to 750 nm for an angular span of 60. For p-polarised emission, the emission peak got tuned from 610 nm to 720 nm for an angular span of 100. Moreover, p-polarised mode showed higher emission intensity as compared to the s-polarised mode. SPCE curves extracted from the image plot between angles 400 and 700 are presented in Fig. (5(c) and polar plot of the same is presented in Fig. 5(d). As in the case of SPCE from silver-based metal films SPCE from gold-based metal films also showed an emission enhancement for thicker dielectric layer. As compared to Au70, where the dielectric layer thickness was only 70 nm, Au270 (dielectric layer thickness ~ 270 nm) showed 400% increase in the SPCE. These results confirm the nature of strong plasmon coupling for metal sensors with thicker dielectric layers is applicable for gold metal sensors as well.
We have also explored SPCE on thicker dielectric layer, sample named Au590. The experimental image plot of the sample Au590 and its simulated SPR image are shown in Fig. 6(a) and Fig. 6(b) respectively. In the simulated SPR image (Fig. 6(b)), we could observe two s-polarised plasmonic modes for Au590: S1 at 420 and S2 at 620. Both exhibited tuning effect, and the later has better tuning from 560 nm to 750 nm for an angular span of 90. Similarly, there are two p-polarised plasmonic modes: P1 at 550 and P2 at 750 also exhibited PCE tuning. On comparing these SPR results (Fig. 6(b)) with the SPCET results (Fig. 6(a)), we could see that the SPCE emission of s-mode and p-mode were integrated to give a single curve. This might be due to the broad emission of both s- and p-polarised emission which appeared closely together due to increase in the thickness of the dielectric and hence an overlap of these peaks resulted into a single peak. But for higher modes, i.e. for S2 and P2, both the modes were well separated in the SPR image and hence the peaks were expected to be resolved in SPCE image. We could see a prominent S2 mode in the SPCE image, but was not able to capture the P2 mode which might be because P2 appears in the higher angular region as shown in Fig. 6(b). The Fig. 6(c) shows the SPCE curves extracted from image plot for the angles between 40o and 70o confirms the tuning of both plasmonic modes. The Fig. 6(d) represents the polar plot of the sample Au590. Apart from dual polarised modes, number of modes got increased with increase in thickness of the dielectric layer. The SPCE intensity of p-polarised emission is more prominent than s-polarised emission. The simulated plot in Fig. 6(b) confirms this observation. Like the sample Au270, a SPCE enhancement of about 300% and emission tuning were also detected in Au590. Thus, a correlation between sample thickness and SPCE enhancement has been confirmed.
Further we investigated SPCE studies on two Cu samples, Cu200 and Cu550. The experimental image plot of the sample Cu200 and its simulated SPR image are shown in Fig. 7(a) and Fig. 7(b) respectively. Unlike Ag and Au samples, dual polarization modes were observed in the SPCE spectra of thin Cu sample. SPCE curves extracted from the image plot are shown in Fig. 7(c). For Cu200, the directionality of SPCE and SPCET were seen in the image plot. The polar plot of Cu200 is given in the Fig. 7(d), which demonstrated the SPCE steering similar to Ag220.
Figure 8(a) and 8(b) represent the experimental SPCE spectral image of the sample Cu550 and its simulated SPR image respectively. SPCE curves corresponding to s and p-polarizations of Fig. 8(b) were observed in Fig. 8(a). The extracted SPCE curves from the image plot are shown in Fig. 8(c). There are three SPCE modes at angular regions: 410 to 440, 460 to 540 and 580 to 650, and they are named as P1, S1 and P2 respectively. Mode Pl shows a PCET from 570 nm to 650 nm with an angular span of 30 and S1 shows a PCET from 580 nm to 760 nm for an angular span of 80. Similarly, the P2 has a PCET effect from 600 nm-720 nm for an angular span of 70. Due to weak plasmon coupling the SPCE due to S2 mode was not observed. On increasing the dielectric thickness, we could observe PCET in Cu samples too as in the case of Ag and Au samples.
The Full Width at Half Maximum (FWHM) values of different SPCET modes of the samples investigated are tabulated in Table.1. Dual polarised plasmonic modes were observed for all metal sensors but are more prominent for metal sensors with thicker dielectric layers as compared to thin dielectric layers. The TWHM of P-polarised modes were observed to be sharper than that of S-polarised modes. FWHM were also compared between sensors with thick and thin dielectric layers. The FWHM values were high for thin dielectric layers but got decreased for thicker dielectric layers. It was interesting to note that there was a decrease of ~ 42% in the FWHM values of P-polarised emission modes for samples with thick dielectric layers as comparison to the samples with thin dielectric layers. Moreover, a 42% decrease in FWHM was found to be a constant in case of all the studied metal sensors. The strong plasmon coupling thus influenced in narrowing the emission by a constant factor of 42%.
Table 1
FWHM values of the SPCET modes (P and S represents P-pol and S-pol respectively and 1,2 are plasmonic mode numbers)
Sl.No.
|
Sample
|
Experimental FWHM (nm)
|
1
|
Ag220
|
71 (P1)
|
-
|
-
|
-
|
2
|
Ag600
|
60 (S1)
|
30 (P1)
|
94 (S2)
|
28 (P2)
|
3
|
Au70
|
85 (P1)
|
-
|
-
|
-
|
4
|
Au270
|
56 (S1)
|
36 (P1)
|
-
|
-
|
5
|
Au590
|
67 (S1)
|
36 (P1)
|
72 (S2)
|
34 (P2)
|
6
|
Cu200
|
37 (S1)
|
90 (P1)
|
-
|
|
7
|
Cu550
|
49 (S1)
|
38 (P1)
|
102 (S2)
|
38 (P2)
|