Photoluminescence is also one of the major features of various properties of quantum dots. When the quantum dots are excited with the external source of energy (photon of light of different energy), the electron in the valence bond or highest occupied molecular orbital gets excited to the conduction band or lowest unoccupied molecular orbital, resulting in the electron-hole pair i.e. exciton. Depending on the type of energy used for the excitation of the different type, this results in different types of photoluminescence, e.g. the photon is used to get photoluminescence, electric field is used for electroluminescence, electron to get cathodoluminescence, etc. (Lide and Mo 1995). These optical absorption energies are evaluated with the electronic structures of the quantum dots. Once the electron is excited to a higher energy level, it will emit its excessive energy via radiative or non- radiative decay, i.e. photon emission or Auger electron emission respectively, resulting in a recombination of the electron-hole pair to reach ground state energy level.
The bright and highly intense photoluminescence of quantum dots originates due to the radiative relaxation of the quantum dots. These radiative emissions may arise due to band edge emission or formation of some defective states in the quantum dots, and are major factors or additional states in quantum dots that give birth to these radiative emissions. This near band emission is basically known to be the exciton energy or electron hole recombination energy that results in the band gap energy of the quantum dots. In the case of defective emission, the photoluminescence is always at a higher wavelength as compared to near band edge emission, due to the intermediate state formed in between the valence and conduction, giving an excited electron an additional pathway for the radiative emission. Details of both processes are explained in the next section.
18.104.22.168.1 Band Edge Emission
Band edge emission and near band edge emission are common processes for radiative emission in the case of semiconductors and insulators. The process of recombination of the excited electron and the hole in the ground state is called a band edge emission. The exciton is bound by a few meV which reveals that the recombination of the electron-hole pair leads near band edge emission which is slightly less than the band gap of the quantum dots. The ground state of the quantum dots is denoted as 1 se—1 sh (exciton state). At room temperature, the full width at half-maxima of near band edge emission of quantum dots lies in between 15 to 30 nm, depending on the diameter of the quantum dots. The photoluminescence can be tailored by the size of zinc selenide quantum dots over the spectrum range 390-440 nm with full width at half maxima as 12.7-16.9 nm (Reiss et al. 2004; Georgios et al. 2004). In the case of bulk semiconductor materials, the photoluminescence emission is easily predictable and quite simply explained by the effective mass model, but at the same time photoluminescence emission of quantum dots is a little bit complicated, e.g. the lifetime of the excited electron in the case of quantum dots is in ns, but in the case of the bulk semiconductors the lifetime is in ps which can be due to the presence of surface states in quantum dots (Bawendi et al. 1990; Efros et al. 1996). The band gap can be either evaluated via absorption spectrum or emission spectrum. In the case of cadmium selenide quantum dots, the photoluminescence shows two emission peaks that originated due to different electron transition centers i.e. lsc-lsh and lse-2sh when observed at a very low temperature (Bawendi et al. 1990). The stoke shift present in the emission is also size-dependent, observed experimentally, e.g. the stoke emission of 5.6 nm cadmium selenide quantum dots and 1.7 nm cadmium selenide quantum dots appears at 2 and 20 meV, respectively. These results can be interpreted via experiments or theoretically that with lowering of the size of quantum dots there is elevation in the distance among the optically prohibited ground exciton state and optical energetic states (Efros and Rosen 2000). In addition to near band edge emission and stoke emission, there are also some irregular emission centers in the quantum dots, i.e. blinking. This term refers to the emission of the quantum dots for some time followed by the dark emission (Nirmal et al. 1996). The reason for the blinking is basically due to the photo-induced ionization process, and further charged quantum dots resulting in the partition between holes and electrons (Efros and Rosen 1997). This theory proposes that the quantum dots show black emission during their ionization state. When cadmium telluride quantum dots were studied as the base model, it was observed the Auger process dominates over the dampening of the photoluminescence emission of ionized quantum dots (Efros and Rosen 1997). The blinking process is not fully reliable with the experimental findings, e.g. experimentally the Auger process displays a linear relationship, while theoretically this process should be linked quadratically to the time of blinking. It is further explored via the Monte Carlo method, when applied theoretically to zinc cadmium selenide quantum dots the blinking process follows the following law, as given in Equation 9.4 (Stefani et al. 2005):
Various other mechanisms have been developed to elaborate the blinking proess but still the precise theory for this model is unknown and there is a lack of proper explanation in the literature (Kuno et al. 2000: van Sark et al. 2002; Stefani et al. 2005; Issac, von Borczyskowski, and Cichos 2005). Carbon dots show UV-visible emission from 290 to 320 nm. This blue shift in the UV-visible absorption appears due to the decrease in the size of quantum dots. Depending on the absorbance of the different sized quantum dots, the emission spectra also changes, e.g. the absorbance at 295, 300, and 325 nm gives an emission at 470, 450, and 383 nm, respectively. Also, the emission spectrum is red-shifted, with an increase in excitation wavelength which is favored by the two mechanisms, i.e. multi-phonon emission and anti-stoke photoluminescence (Sun et al. 2013). The photoluminescence emission of mostly studied carbonaceous quantum dots is endorsed for quantum confinement and/or the surface defects developed in the carbonaceous quantum dots. Still, the origin of photoluminescence emission of carbonaceous quantum dots remains unclear (Xu et al. 2014). One report on the photoluminescence emission of carbon dots reveals that the bright emission originates due to the quantum confinement of the graphitic structure, and not due to the defective emission (Li et al. 2010).
22.214.171.124.2 Defect Emission
Radiative emission other than near band edge emission often originates due to various reasons; impurities or defective states in the crystal lattices is one of the major contributors to photoluminescence emission (Issac et al. 2005). These defects may be either donor or acceptor, depending on the type of impurities, i.e. excessive electron and electron-deficient. These defective states exert Coulombic attraction to the exciton pair, i.e. the hole is attracted towards the donor type defect, and the electron is attracted towards the acceptor type defect. When these types of interactions are modeled on the hydrogen atom, this results in the decrease of binding energy and approaches to the dielectric constant of the materials (Gfroerer 2006). These defective states are of two types, i.e. shallow states that lie in the conduction band of the nanocrystal while the other, i.e. the deep state, lies in the valence band. The shallow defect displays radiative emission and effective generally at a low temperature because at room or elevated temperature the exciton gains energy and leaves the defective states. In the case of deep level defective states, the excitons are long-lived (high life time) and display typically non- radiative emission. Emissions from these two defective states are dependent on the concentration and are helpful in examining their energy. The photoluminescence emission is the combined contribution of both band gap and defective emission, as on tailoring the excitation energy both the distribution and intensity of the emission spectra changes. In addition, the excitation wavelength is also helpful in evaluating the photoexcited state of quantum dots but the lifetime of these is very short (Gfroerer 2006). Due to the large volume-to-surface ratio, these defective states are also expected to be on the surface of the quantum dots and are generally called surface defects. Various methodologies have been developed to improve the surface and make it defect- free, but still there is always a lack of one hundred percent defect free surface of the crystals (Cheng et al. 2006). These defective states arise due to synthesis processes and/or passivation of the surfaces. These surface defects damper the luminescence as well as the electrical property of the quantum dots via non-radiative emission (Djurisic et al. 2004). Yet there are some quantum dots where the surface defect leads to highly intense radiative emission, e.g. zinc oxide. In fact, zinc oxide nanocrystals are a perfect example to understand the surface defects, donor, and acceptor defects. In addition, the photoluminescence emission, as well as the size control of zinc oxide nanoparticles is very much dependent on the solvent system. When water is taken as a solvent, various defective emissions originate in the crystals which are elaborated below. Mishra et al. have synthesized zinc oxide nanocrystals and decoded the origin of peaks in the photoluminescence spectrum (Mishra et al. 2010). The excited electron dissipates via five different pathways, as shown in the photoluminescence of zinc oxide. Emission appears at 396 nm and originates due to the near band edge emission, i.e. exciton peak (Vanheusden et al. 1998). In addition to peak at 396 nm, i.e. 416, 445, 481, 524 nm, it contributes to the emission spectrum, and originates due to the presence of various defective states in the crystals, such as oxygen vacancy (V(v), zinc vacancy (VZn), interstitial oxygen (O;), interstitial zinc (Zn), and oxygen antisites (0Zn). A peak centered at 416 nm appears due to the electronic shift from the donor level of Zn, to the valence band (Fan et al. 2005). Luminescence appears at 481 nm, and is due to the contribution of radiative transition of the shallow donor level to Znj to a higher acceptor level of neutral Zn (Tatsumi et al. 2004). Different theories have been developed by various research groups to describe the origin of the final peak, i.e. that at 524 nm. Dingle, in his experiments, reveals that the origin of this peak depends on the presence of a trace amount of copper ions (as impurity) in the lattice structure of the zinc oxide (Dingle 1969). Similarly, another theory explained the presence of oxygen deficient environment that gives rise to the vacancies of oxygen and hence leads transition between V0-acceptor VZn, which is responsible for the emission of the peak at 524 nm (Heo et al.2005). In addition to these two theories, another is also present in literature for the origin of the peak at 524 nm and is highly cited - that explains the origin of the green emission due to the presence of interstitial oxygen in the crystal lattice (Wen et al. 2005). The detailed scheme of different photoluminescence emission centers of zinc oxide nanocrystal is illustrated in the scheme given in Figure 9.5.
Liqiang and his group have explained the effect of diverse excitations on the luminescence spectra of zinc oxide (Liqiang Jing et al. 2005). The PL emission ranges from 400 to 500 nm have two intense peaks at 420 and 480 nm, characteristically due to the band edge free and binding excitons, respectively (Zhang et al. 2003; Lide and Mo 1995). Interestingly, when the reaction temperature increases, these defective states diminish and the defective emissions merge into a single peak except near band edge emission (Tam et al. 2006). By increasing the annealing temperature, the defective peak intensity increases and the shape of the emission peak also changes. When the temperature is elevated from 400 to 600 °C in air, both orange and green emission of zinc oxide appears. In addition to the reaction temperature, the reaction atmosphere also plays an important role. At 200 °C, in the presence of forming gas (N, and H2), there is no signal of defective emission of zinc oxide (Figure 9.6). When the temperature is elevated from 400 and 600 °C in an inert atmosphere, green emission with a strong intensity appears.
From above, it is clear that in the aqueous system there is a distinct photoluminescence emission of zinc oxide but the intensity of these emissions is quite low - in fact the main disadvantage is the hydrolysis of Zn2+ ions to Zn(OH)2 and Zn(OH)42~ is very fast and basic. Therefore, controlling the photoluminescence as
FIGURE 9.5 Origin of emission peaks in ZnO nanoparticles.
well as the growth of nanocrystals in an aqueous medium is a very tedious process. Moving further, Hu et al. have studied the effect of various non-aqueous solvents on the PL emission of zinc oxide nanoparticles (Hu et al. 2010). Figure 9.7 depicts the photoluminescence emission of zinc oxide nanocrystals in different solvents showing intense and broad green emission. Near band
FIGURE 9.6 PL spectra of ZnO nanorods at different annealing temperatures for samples annealed in: (a) air, (b) forming gas, different annealing atmospheres for annealing at (c) 200 °C, and (d) 600 C. (Source: Tam et al. 2()06)
edge emission disappeared in the case of methanol and acetone while other solvents show a distinctive near band edge emission at 390 nm. In addition, the photoluminescence emission intensity increases with increases. The intensity and position of the photoluminescence emission of zinc oxide nanoparticles increased with an increase in the size of nanoparticles and by changing the solvent respectively. Actually, the excessive hydroxy groups present are due to the basic solution on the surface of zinc oxide nanoparticles which give rise to the surface defects during the growth process. The larger surface area and more surface defects give rise to lower ultra violet or near band edge emission of zinc oxide nanostructures.
In the case of carbonaceous quantum dots, the origin of photoluminescence is still a mystery as explained above. When the reason for photoluminescence emission is explored, there is other published literature that reveals excitons of carbon, aromatic systems, quantum confinement, trapped states, defects dues to edges, oxygen-containing groups, and zigzag sites contribute to photoluminescence emission. Bao et al. observed electro- chemically that irrespective of the particle size, the increase or decrease in wavelength happens due to the surface of the carbon dots (Bao et al. 2011).
126.96.36.199.3 Activator Emission
Photoluminescence emission of quantum dots also arises due to the incorporation of impurities, which is known to be extrinsic photoluminescence. The mechanism that favors extrinsic photoluminescence is exciton transition which can occur between the donor to valence state or the donor to acceptor level or the conduction to acceptor state. In certain cases, this emission center is localized on the activator atom center. In the majority of cases, the rule for the emission is hassle-free in ligand or crystal field, due to orbital’s intermixing, such as d-p where the orbitals are fragmented into hyperfine splitting. In addition, d-d transition is allowed in the case of transition metals (Yang et al. 2004). When manganese (II) is doped in cadmium sulfide and the surface is passivated with zinc sulfide, the increase in the lifetime of the excited electron is of ms which arises due to forbidden
FIGURE 9.7 PL spectra of the ZnO nanostructures in (a) methanol, (b) ethanol, (c) 1-propanol, (d) 1-butanol, (e) 1-pentanol, (f) 1-hexanol, (g) acetone, and (h) isopropanol solutions of NaOH, respectively. (Source: Adapted from Hu et al. 2010)
d-d transition of manganese (II) impurity (Yang et al. 2004). Similarly, when f block ionic impurities are added to the quantum dots structures, there is often f-f transition which is examined in the photoluminescence emission (Lee 1996). The f level impurities are isolated by the crystal field of the host crystal lattice via shielding of exterior p and s orbitals, resulting in the emission similar to atomic spectra. A lot of work has been done on the extrinsic optical emission of doped zinc oxide quantum dots that include transition as well as rare earth elements (Xiu et al. 2006; Ranjani Viswanatha et al. 2006; Wang et al. 2006). When zinc oxide quantum dots have been doped with terbium ions, two types of emission have been observed in the photoluminescence emission: one corresponds to terbium ions, and the other corresponds to the defective state (Liu et al. 2001). When dopant concentration increases, the emission of terbium increases, while at the same time the emission peak of defects decrease. Nevertheless, when zinc oxide is doped with dysprosium ions, it exhibits a relatively strong ultra violet emission with a very weak emission from dysprosium (Wu et al. 2006). The emission of manganese-doped zinc oxide quantum dots depends strongly on the reaction condition (Zhang et al. 2003). The manganese- doped zinc oxide quantum dots leads to dampening of the green emission, as well as other defective peaks in the crystals; at the same time some results also report the blue shift, as well as an increase in the intensity of the near band edge emission. Zinc sulfide based-doped quantum dots are also a very important class of semiconductors. Manganese-doped zinc sulfide quantum dots are a highly explored combination for phosphor. When zinc sulfide is doped, the manganese results in the enhancement of the quantum yield, as well as an emission intensity of the zinc sulfide quantum dots with ns of the lifetime of the excited electron (Bol and Meijerink 1998; Su et al. 2003). The reason for the increase in intensity is accredited by the effectual energy transfer from the zinc sulfide quantum dots to the manganese (II) ions, enabled via the mixing of the orbitals. The intermixing of the orbitals of zinc sulfide and manganese also attributes to the short lifetime of the excited electron. The effect of nitrogen doping on the photoluminescence emission of carbon dots has also been explored by Gong et al. to explore the spectral and luminescence properties of carbon dots (Gong et al. 2015). The emission peak bare carbon dots remain constant when excited from 280-340 nm. however when the excitation wavelength increases from 360-480 nm, there is a red shift in the emission spectrum of carbon dots. A similar type of emission results was obtained for nitrogen-doped carbon dots. The independent nature of the emission spectra on excitation is due to the core л->я* transitions of graphitic nature, while the excitation-dependent nature of the quantum dots is due to the surface defective state which arises due to the functionalities present on the surface of the quantum dots (n—>я:*). In addition, the photoluminescence emission intensity of nitrogen- doped carbon dots is much higher as compared to the bare carbon dots, revealing the higher density of surface defects on the surface of carbon dots due to various functionalities of nitrogen. In another report, when carbon dots are doped with nitrogen and sulfur, nitrogen-doped carbon dots show a higher quantum yield as well as the lifetime of the excited electron as compared to sulfur-doped, as well as the bare carbon dots. This is due to the reason that the nitrogen-doped carbon dots fabricate a new kind of surface defective state, giving new a pathway for the excited electron to radiative decay (Wenjing Lu et al. 2015). The photoluminescence emission of nitrogen-doped carbon dots can be tailored from blue, green, and yellow via the reduction of the concentration of doping of nitrogen, and is similar to the result obtained from graphene quantum dots (Wu et al. 2014; Tetsuka et al. 2012).
Quantum Yield of Quantum Dots
The measurement of the quantum yield is vital for quantum dots and is also an important property of the quantum dots. It has been observed that the quantum yield values differ for particular quantum dots among different reports. There may be several reasons that include: a different approach to measure quantum yield; a difference in the concentration of quantum dots or sample or both; alteration of the slit width for measurement; instrumental error may be there (van Sark et al. 2002). The inorganic semiconductors may not perform as dye, but in the case of carbonaceous quantum dots, they behave somewhat similarly to dye in terms of quantum yield. The methodology to determine the quantum yield is by relating the emission intensity of the quantum yield with that of the standard taken, followed by the measurement of the absorbance of the quantum dots as well as the standard, with one condition that the absorbance value of both the quantum dots and standard are kept below 0.08. The following relation was used to determine the quantum yield of the quantum dots (Bera et al. 2008a: Qian et al. 2008):
In the above equation, std refers to standard, q refers to refractive index, I stands for emission intensity. When measuring the intensity for standard and sample, the excitation wavelength should be the same.
Dai et al. (2018) have synthesized cadmium selenide quantum dots that have been shown to be the quantum yield of about 64% using rhodamine 101 in ethanol which is shown to be stable for up to 120 days (Dai et al. 2018). When core shell quantum dots have been prepared, i.e. cadmium selenide/zinc sulfide, the quantum yield is hampered to the value of 50%, with the aid of rhodamine 560 in ethanol solution (excitation 560 nm, emission 480-850 nm). In another example, when the zinc sulfide is replaced by cadmium sulfide, i.e. cadmium selenide/cadmium sulfide, there is improvement in the quantum yield (84%) (Xiaogang Peng et al. 1997). A 50% quantum yield has been observed for zinc selenide emitting blue light using stilbene as the standard and methanol as solvent (Margaret and Guyot-Sionnest 1998). When the zinc selenide is doped with manganese, there is dampening of the quantum yield i.e. 22% (Norris et al. 2000). It has been reported in the literature that the composition as well as the stoichiometric ratio of the precursor are significant areas to improve the quantum yield of the quantum dots (Qu and Peng 2002). When the particle size of the zinc oxide quantum dots is tailored from 0.7 to 1 nm, there is a decrease in the quantum yield of quantum dots (van Dijken et al. 2001). Quantum dots that are appropriately surface- passivated have an elevated quantum yield in the visible region (400-700 nm), e.g. cadmium selenide (65-85%), cadmium sulfide (60%), and indium phosphide (10-40%). The quantum yield of carbonaceous quantum dots are carbon dots and nitrogen-doped carbon dots are 5.6 and 15.8% (Wu et al. 2014). Sun et al. have synthesized nitrogen-doped graphene quantum dots with excellent water stability that show an enormously high quantum yield of about 74% with the aid of rhodamine В (Sun et al. 2015).