Fig.2 shows the XRD patterns of HgI2 pellet and nanoparticles. The XRD pattern of bulk HgI2 exhibits eleven peaks at 2θ = 14.50, 21.70, 250, 29.70, 32.40, 35.40, 41.20, 43.80, 47.10, 48.80, and 59.90 related to (002), (101), (102), (103), (112), (104), (114), (006), (211), (106), and (008) planes, respectively. All these observed peaks are related to tetrahedral structure of HgI2 planes according to JCPDs #73-0455. The XRD patter of HgI2 NTs shows the presence of five peaks positioned at 2θ = 130, 21.70, 25, 39.10, and 53.10 corresponding to (002), (101), (102), (123), and (214) plane, respectively. The (002), (123), and (214) plans are belong to orthorhombic β-HgI2 according to JCPDS-ICDD #15-0034, while (101) and (102) plans are indexed to tetragonal α-HgI2 [7,10], indicating that the synthesized HgI2 NTs are mixture of two phases α-HgI2 and β-HgI2. As shown, the highest peak intensity for HgI2 NTs was observed for (002) plane confirming that the grains grow in the direction perpendicular to the substrate (c-axis). The average crystallite size (D) was determined from XRD data with using Debye–Scherrer equation along (002) plane
Where λ is the wavelength of X-ray source and the β is the full width at the half maximum in units of radians. Our calculation revealed that the average crystallite size of HgI2, dislocation density and the strain were 14nm, 5.32×10-3nm-2, and 25.3×10-3, respectively.
The optical absorbance of HgI2 colloids is shown in Fig.3. As shown, the optical absorbance of HgI2 was flat in the range 250-350nm and it is sharply decreased after 347nm and saturated after 400nm.
The optical energy band gap was calculated using Tauc plot. As shown in Fig.4, plotting of the (αhѵ)2 vs (hѵ) and extrapolating the linear part of the second region to the photon energy gives the optical energy gap which also indicating that HgI2 is direct band gap [11]. The direct optical band gap of HgI2 NTs was found to be around 2.9eV.
Fig.5. shows the zeta potential (ZP) plot of HgI2 colloids which is approximately 22mV confirming the synthesized HgI2 NTs have high degree of stability and high desperation with low tendency to for aggregation and agglomeration [12,13].
Fig.6. demonstrates the PL spectrum of HgI2 colloids exited with laser source of 300nm wavelength, the spectra show the existence of only single PL emission centered at 535nm (2.31eV). This value is smaller than the optical energy gap determined from UV-VIS results due to the effect of trapping defects inside the band gap of HgI2 [14].
Fig.7 shows the IR vibration spectrum of HgI2 colloids Two IR assignments were observed at 612 and 1102 cm-1 that belonged to Hg-I and the C-O stretch, respectively [15].
The EDX spectrum of HgI2 deposited on silicon substrate is depicted in Fig.8 which is showed the existence of mercury and iodide elements and we attributed the presence peaks of carbon and silicon to the organic environment and the substrate, respectively [16]. The Hg/I weight ratio was 1.23 indicating the synthesized HgI2 NTs were small-off stoichiometric (see the inset of Fig.8).
Element
|
Weight
|
I
|
27.13
|
Hg
|
33.62
|
Si
|
32.56
|
C
|
4.61
|
Fig.9. illustrate SEM image of HgI2 with two magnifications. Spherical nanoparticles mixed with nanotubes were observed (Fig.9-a). Fig.9-b shows the SEM image of tetragonal bi-pyramids or truncated pyramids structure of HgI2. The particle size distribution given in Fig.9-c confirmed that mean particle size of HgI2 was 23 nm. We have observed the presence of only monodispersed nanoparticles which in good agreement with results of Zeta potential result. Fig.10 shows the TEM image of HgI2 which confirmed formation of monodispersed nanotubes with average diameter of 80nm and average length of 1.6μm as well as the existence of spherical nanoparticles. As shown in TEM image, some of the nanoparticles are found to attached to the nanotubes.
Fig.11 shows the current-voltage properties of HgI2 NTs/Si heterojunction measured at room temperature at. It is clear that the forward current consisting of two regions; in the first region the current increased slightly with bias voltage with which indicates that the recombination current is larger than generation current, while at large bias the current increases exponentially with voltage due to domination of diffusion current [17-19]. The reveres current was shown to be increased slightly at small voltage and after bias with 2V, it increased significantly due to the surface leakage current flow through the edges of the heterojunction.
The ideality factor of the heterojunction was calculated using diode equation and its value was 4 for first region and 12 for second region. When the ideality factor > 1 indicates that there are structural defects and surface states in the interface region [ 20] as well as due to the mismatch in lattice constants between HgI2 and silicon. As the photodetector illuminated with white light, the current of the photodetector increases due to the production of e-h pairs in the depletion region. Fig.12 depicts the illuminated I-V characteristics of HgI2/p-Si heterojunction under white light illumination. The photocurrent increases with bias voltage as a result of widening of the depletion region. The photocurrent increased with light intensity, this increasing attributed to more photon absorption and generation of electron-hole pairs in the depletion layer [21]. The linear dynamic range (LDR) of the photodetector was determined from variation of photocurrent with light intensity plot shown in Fig.13. The saturation in photocurrent was started at light intensity of 180mW/cm2.
The most important figure of merit of the silicon photodetector is responsivity which is defined as the ratio of the generated photocurrent to the power of incident light. Fig.14 illustrates the responsivity versus wavelength of HgI2/p-Si HJ photodetector measured at 5V bias. We can see that the maximum responsivity was 1.09AW-1 at 450nm which is higher and comparable to some large band gap heterojunction-based silicon photodetector as shown in Table 1. We attributed the improvement in the responsivity at visible region to the large surface area of nanostructured HgI2 that comes from nanotubes morphology [22].
Table 1. Figures of merit of present work compared with other photodetectors
Photodetector type
|
Responsivity (A/W)
|
Detectivity (Jones)
|
Quantum efficiency%
|
HgI2 NTs/Si [present]
|
1.09 at 450nm
|
3.6×1012 at 400nm
|
3×103
|
CdTe/Si [23]
|
0.5 at 950nm
|
1.2×1011 at 950nm
|
65 at 950nm
|
CdO/Si [24]
|
0.5 at 600nm
|
7×1011 at 600nm
|
62 at 600nm
|
CdS/Si [25]
|
0.59 at 1064nm
|
1.3×1012 at 1064nm
|
-
|
The responsivity of the photodetector at 450nm is due to the absorption edge of HgI2 NTs which matched with result of band gap. Fig.15 show the detectivity and quantum efficiency spectra of HgI2/Si HJ photodetector. The maximum detectivity was found to be 3.6 ×1012 Jones at 400nm wavelength. The of HgI2/Si HJ photodetector show high quantum efficiency about 3×102 at 450nm. This high quantum efficiency gives an indication the high value of carrier collection efficiency [26-28].