Switchable Multi-Color Solution-Processed QD-laser

In this paper, for the first time, the switchable two-color quantum dot laser has been realized considering solution process technology, which has both simultaneous and lonely lasing capability exploiting selective energy contacts. Furthermore, both channels can be modulated independently, which is a significant feature in high-speed data transmission. To this end, utilizing superimposed quantum dots with various radii in the active layer provides the different emission wavelengths. In order to achieve the different sizes of QDs, solution process technology has been used as a cost-effectiveness and fabrication ease method. Moreover, at the introduced structure to accomplish the idea, the quantum wells are used as separate selective energy contacts to control the lasing channels at the desired wavelength. It makes the prominent device have simultaneous lasing at different emission wavelengths or be able to lase just at one wavelength. The performance of the proposed device has been modeled based on developed rate equation by assuming inhomogeneous broadening of energy levels as a consequence of the size distribution of quantum dots and considering tunnel injection of carriers into the quantum dots via selective energy contacts. Based on simulation results, the simultaneous lasing in both or at one of two wavelengths 1.31  μm and 1.55  μm has been realized by the superimposition of two different sizes of InGaAs quantum dots in a single cavity and accomplishment of selective energy contacts. Besides, controlling the quantum dot coverage leads to managing the output power and modulation response at the desired wavelengths. By offering this idea, one more step is actually taken to approach the switchable QD-laser by the simple solution process method.

. The schematic and performance of the proposed switchable two-color QD-laser utilizing SECs. (A) The block diagram of the proposed switchable two-color QD-laser. (B) The structure of two-color QD-laser with superimposing two different sizes of InGaAs QDs in active region and two SEC corresponding to each QD groups located between SCH layer and active region to tunnel injection of carriers directly into the related QD groups with the same energy levels. (C) The energy band diagram of the proposed structure including the carrier relaxation process into QDs.

the proposed switchable two-color QD-laser using Secs
As stated above, in this paper the switchable two-color solution-processed QD-laser exploiting SECs has been proposed by the ability to choose lasing wavelengths. In this QD-laser, the quantum well (QW) 44,54 is used as the SEC to control the lasing channels. Here, the most usual 1.31 μm and 1.55 μm wavelengths in the telecommunication applications are considered. As seen in the schematic view of the device structure in Fig. 1(B), for the realization of two-color lasing, two different sizes of InGaAs/GaAs QDs with radiuses of R 1 and R 2 are superimposed in the active region of QD-laser, so the lasing from two independent channels becomes possible. For facile access to this structure, the simple and low-cost solution process method is suggested, which is called "wet, " and it provides the chemical system with controllable experimental conditions. The synthesis of colloidal QDs by chemical procedures in solution allows achieving tunable sizes of QDs 30,55 ; also the creation of suitable conditions for synthesis in this method is more convenient. The solution process is economical compared with the other expensive methods like epitaxial which needs a high-vacuum condition 30,31 .
Injection of carriers into selected groups of QDs can be realized by SECs. Thus, the width of the QW is chosen as their energy state is the same with the GS energy of relevant QDs; then tunneling injection of carriers occurs via the barriers of QW into QDs, so they get localized in them. This process causes the fast transition of carriers 30 and improves the performance of QD-laser. Figure 1(B) exhibits the two-color QD-laser structure in which the active layer of QD-laser includes two different sizes of InGaAs QDs that is sandwiched by p and n-type AlGaAs as a carrier transition layer. The p and n-type GaAs are cladding layers. Also, the QW contacts contain the InGaAs well and AlAs barriers. As seen in the figure, current injection in each of the contacts can active them, therefore if the current is injected in both of the contacts, the output emission occurs in two colors, but the injection of current just in one contact leads to having one color emission; note that the key idea is the operation of SECs to control the output emissions, accordingly the title of switchable two-color QD-laser by selective contact can be chosen for this structure. The process of the system starts with the diffusion of carriers in the SCH layer by current injection after that the carriers relax into the QW SECs. The collection of carriers, tunnel from QW state into GS of QD, and then the transmitted carriers recombine radiatively or non-radiatively. Figure 1(C) illustrates the carriers act in the energy band diagram of switchable two-color QD-laser by SEC. Obviously, the big QD that determines the channel1, has both GS and excited state (ES), whereas the small QD relates to channel2 with only GS. The carriers and photons dynamics like capture and relaxation in the QD describes the performance of QD-lasers. As it is shown in detail in the figure, at first, the diffusion of the carriers occurs in the SCH layer (τ si ) and then some of those carriers tunnels into the QDs (τ tun,i ). These carriers recombine radiatively or non-radiatively; the lifetime constant of recombination of diffused carriers in the SCH layer and in SECs are τ sr and τ sc,r , respectively. In the big QDs, the carrier relaxed from ES to GS by the rate of τ E→G ; also some of the carriers re-excited from ES to GS by the rate of τ G→E . Moreover, the non-radiative recombination of carriers in the QDs energy levels determine by τ Er and τ Gr . Finally, the stimulated emission leads to the photon emission from the ground state by the rate of τ p .

concept and theoretical Modeling
In this section, the theoretical model is provided for the proposed structure. As mentioned before, the aim of this paper is achieving to a switchable two-color QD-laser by selective contacts. According to reports, some of the research groups have suggested the InGaAs QDs for QD-laser structure 20,23,56 . Therefore, utilizing the two QDs of InGaAs/GaAs with radiuses of 2.47 nm and 3.85 nm in the active region leads to achieving two colors at an independent emission wavelength of 1.31 μm and 1.55 μm, respectively. Even though the solution process is a low-cost method which provides a simple chemical system to achieve controllable experimental condition to synthesize different sizes of QDs 38 , However, due to the insignificant limitations of synthesizing condition, the exact size of QDs can't be attained; thus the energy levels of QDs is broadened inhomogeneously as a result of the size distribution of QDs 20 during the solution process method. Generally, as it is seen in Fig. 2(A), the IHB of energy levels has to be considered which is modeled by Gaussian function 20,23,53 . The FWHM of the Gaussian function can take specific values as Γ G . In addition, the effect of carrier-carrier and phonon-carrier scattering leads to the homogeneous broadening (HB) of energy levels which is presented by Lorentz shape 20 .
One can realize from Fig. 2(B) that E-R relates to energy versus QD size; also, the suitable radius of QDs shows the resonant energies corresponding to desired emission wavelength; (0.8 eV and 0.946 relate to 1.55 μm and 1.31 μm, respectively). The diagram of Fig. 2(B) is obtained as the consequence of numerically solving the Schrödinger equation for switchable two-color InGaAs/GaAs QD-laser. As it is shown in Fig. 2(B), proportionate www.nature.com/scientificreports www.nature.com/scientificreports/ to introduced Γ G , there is size distribution over both specific R 1 and R 2, each of which is denoted by ∆R 1 and ∆R 2 , respectively; note that ∆R 1 is broader than ∆R 2 . Following this result, achieving an efficient device is possible by the solution process method which provides high accuracy experimental conditions to synthesis a small radius of QD with 0.5 nm deviation and the big one with 0.1nm 40 . As reported, the performance of QD-laser is based on rate equations 20,24,40 , so the analyzing proposed structure in this research by consideration of both IHB and HB of QD's energy levels is carried out based on developed rate equations. Selective low coverage of QDs in the structure causes the QDs not to have correlation and become isolated from each other 40 .

Modeling Switchable Two-Color QD-Laser in Rate Equation Framework
According to the last reports, solving the rate equations is the most popular way to analyze the QD-laser dynamics 20,24 . In this model, the rate equations are developed to calculate the characteristics of a two-color QD-laser, including SECs. Needless to say, the QDs without any correlation are isolated from each other and it is supposed that both the electron and hole are captured in pairs at GS of the QDs, which form an exciton; in addition, the escape of carriers from SCH layer is neglected, and the following set of developed rate equation is used just for electrons. Here, we assume two ensembles of QDs, each of which contains n = 1, 2, …, 2 M + 1 groups corresponding to the desired wavelengths. The energy width is described as the mode separation of each group and is given by where c and h stand for the speed of light in vacuum and Planck's constant respectively, and L cav is the length of the cavity. Due to the resonance of all modes, L cav is selected relevant to the appropriate case. Therefore, L cav equals to the multiple of the cavity length of each group 40 , L L a n a n a n 2 2 (2 ) where L cav1 and L cav2 correspond to the cavity length of first and second groups of QDs' ensemble. The cavity length should be selected in a way that the coefficients a, a 1, and a 2 become integer. It is good to be mentioned that in this equation, only the value of each wavelength is considered and the calculated L cav is expressed in μm 40 . Considering solution process technology, the size of each QD group relates to the emission wavelength of each channel and can be deviated from the value R 1 or R 2 , leading to the broadening of energy levels in each QD ensembles. The IHB of energy levels can be modeled by the Gaussian function 20 around the central interband transition energy of each channel, E 1 and E 2 , which is defined as  www.nature.com/scientificreports www.nature.com/scientificreports/ Also, the FWHM of IHB equals to ξ Γ = .
2 35 35 G 0 . The energy E n i , is given by,

The main mode of each group is defined as
, that relate to 1.55 μm and 1.31 μm, respectively. In addition, the effect of temperature and various scattering mechanisms including carrier-carrier and phonon-carrier scattering leads to HB, which has a Lorentzian lineshape function 20 as defined by According to B i,mn the FWHM is 2ℏΓ B in which Γ B is the polarization dephasing or scattering rate 20 . In this equation E m 1, , E n 1, (E m 2, , E n 2, ) refer to the energy of mth mode and nth group of big QDs, relevant to channel1 (small QDs corresponding to channel2). The QD laser's linear optical gain is based on density matrix theory and it is represented by where n r is the refractive index and N N ( ) 2 . The sizes of R 1 = 3.85 nm and R 2 = 2.47 nm determine the lasing wavelength at 1.55 μm and 1.31 μm, respectively. When the HB of energy levels is small in comparison with IHB, the different sizes of QD groups have been spatially isolated from each, leading the optical gain of lasing modes independent of other modes. In better words, when the HB is approximately comparable to IHB, the resonant and non-resonant modes whose energy levels lie within the amount of HB have been considered in the optical gain of lasing modes. The occupation probability in the GS and ES of the QD's ensembles is in accordance with Pauli's exclusion principle which is defined by GSi n GSi n where N GS1 and N GS2 are the carrier numbers of GS for channel1's QDs and channel2's QDs, respectively; also N ES1 is the carrier number of ES of channel1's QDs. The active region's volume is V A , and the QDs' densities of each channel are given by N N , In the optical gain formula, P i 2 is the transition moment matrix element which is defined as According to the above formula, I cv defines the overlap integral between the envelope functions of both a hole and an electron for each channel's QDs. M b is got from k.p theory with the first-order interaction between the conduction and valence bands. The transition matrix element is given as follows, in which E g is the bandgap, Δ is the spin-orbit, and ⁎ m e is the electron effective mass 20 .
According to defined relations, as stated above, the rate equations of switchable two-color QD-laser by SEC are expressed by    are the photon number of mth mode of QDs corresponding to channel1 (channel2). The spontaneous emission coupling efficiency is defined by β. The lifetime of injected carriers from the SCH layer into SEC is defined by τ si , and the τ tun1 ; the tunneling time of injected carrier from SEC into relevant GS level of QDs is τ tun2 and it will be explained in the next section. Diffused carriers are recombined in the SCH layer with τ sc,r . The carrier recombination in the SEC is given by τ sr . The τ Gr , τ Er are non-radiative recombination of carriers in the QDs energy levels. The carrier escape rate from the SCH layer is τ sc,e . The chan-nel1's QDs contain two states (i.e. GS and ES); so the carrier relaxation rate from ES to GS and carrier re-excitation from GS to ES in the big QD that are defined by E G τ → and τ → G E as follow 23 are the initial relaxation and re-excitation rates when the GS and ES are empty, respectively. The initial relaxation and re-excitation time are related by 23 − ) is the energy difference between GS and ES in big QD. The photon emission of each channel is caused by the radiative recombination of carriers in both big and small QDs, and the photon lifetime of the cavity is defined by τ p as follow 20 c n L r r where r 1 , r 2 are the cavity mirrors reflectivity and α i is the internal loss. As mentioned above, the advantage of the proposed structure of this paper is the possibility of lasing at two emission wavelengths from both channels simultaneously or individually; therefore, the output power can be calculated from independent channels. Note that the evaluated output power of the laser is given from one of the mirrors as follow here the emitted photon energy from channel1 is E 1,m with the main energy of 0.8 eV; also, E 2,m equals to 0.946 eV as the main energy of channel2. The mirror reflectivity is r, which can be substituted with r 1 or r 2 .

Selective energy contacts
Achieving to switchable two-color QD-laser is possible by using SECs, which can inject the electrons into the certain QDs. In addition, coupling the QW leads to the improvement of the fast transition of the carrier 44 . In this model, utilizing two different sizes of QDs in the active region of QD-laser results in two-channel emission wavelengths while the lasing from each channel can be controlled by the independent SECs as an injection of carriers. Effective injections of carriers collection from SECs to QDs provide enhancement performance of the structure. The transition of the carriers through QW to QDs is defined by tunneling rate, which is controlled by the material parameters and thickness of the barriers 45 . Also, the thin enough barrier layers lead to effective tunneling through the SECs to QDs. Here, the QW and QDs are selected InGaAs, where the barriers are AlAs. Also, the energy levels of the QWs are proportional to QD's GS, i.e.; the thickness of the SEC should be selected in a way that the energy levels of SEC and QD's GS are equal.
The modal analysis of a switchable two-color QD-laser has been done by solving the Schrödinger equation using the Finite Element Method (FEM). The Eigen-energy and corresponding Eigen-function of the system and also the tunneling rate has been calculated. It is good to be mentioned that continuous Periodic Boundary Condition (PBC) is selected for boundaries along the x-axis for active region and SECs. The Eigen-energy of structure respect to the conduction band edge is illustrated in Fig. 3(A). The orange and green solid line corresponds to big and small QDs, which is coupled to relating SECs. The transmission rate of injected electrons from each SEC to QDs along the y-axis has been calculated and depicted in Fig. 3(B). As shown in this figure, there are two peaks corresponding to the main mode of each QD group at 0.8 eV and 0.946 eV. The wavefunction of each QD group coupled to relevant SEC has been shown in Fig. 3(C),(D) for channel1 and channel2, respectively. According to Fig. 3(C),(D) the electrons injected from the right (left) SEC coupled to big (small) QDs and Scientific RepoRtS | (2020) 10:5273 | https://doi.org/10.1038/s41598-020-60859-w www.nature.com/scientificreports www.nature.com/scientificreports/ tunneled between QDs. The tunneling time τ tn i , is defined as follows, which T trans is the transmission rate of carriers and N SDi is the surface density of QDs ensemble 44 . τ − and it is given by

The Dynamic Behavior of Switchable Two-Color QD-Laser
The dynamic behavior of the two-color QD-laser has been simulated by the calculation of the small-signal modulation response 26,51,57 . To this end, a current step has been applied to each SECs as follows where U t ( ) is a unit step function, I 1 and I 2 are the bias current of channel1 and channel2, respectively. The bias currents have to be larger than the threshold current of the related channel. The ΔI i (ΔI 1 and ΔI 2 ) is the small step perturbation which is much smaller than threshold current, applied to each channel. The duration of the current step has to be long enough to ensure that the transient response settles down. The improved rate Eqs. (11)(12)(13)(14)(15)(16), taking into account the modulated current for each channel mentioned in Eq. (24), have to be solved based on the fourth-order Runge-Kutta method. The small-signal modulation response of a two-color QD-laser can be obtained by the Fourier transform of photon densities considering the small step current.

The Switchable two-Color QD-Laser utilizing SECs
The switchable two-color QD-laser using SECs has been characterized by solving the improved rate Eqs. (11)(12)(13)(14)(15)(16) based on fourth-order Runge-Kutta method, in order to time evaluation of carriers and photon densities. The output power emitted in accordance with the biasing of each channel can be obtained through photon densities at a steady-state. Hence, the lasing spectra relevant to each channel of the proposed QD-laser have been calculated www.nature.com/scientificreports www.nature.com/scientificreports/ on the steady-state output power of each main mode. Moreover, calculating the steady-state carrier densities in GS of big and small QDs leads to obtaining the gain spectra relating to each channel.
In the suggested model, the re-excitation of carriers are ignored; we assumed at first all the carriers are injected in the SCH layer, then those tunnel from the contact to the QDs; in this assumption, the escape of the carriers from SCH layer is neglected. The utilization of QDs in different sizes, observed in this type of QD-laser, is provided by solution process technology, which is a simple and low-cost method. In this process, the simple condition of the chemical system permits to achieve tunable sizes of QDs that are used in the high-performance devices. In the solution process, the QD size deviation of around 0.1 nm can be possible in the appropriate experimental condition of the chemical procedure, which causes the broadening of origin energy. Therefore, the size fluctuation of QDs always occurs in the QDs ensemble, and efficient tunneling will be done within a certain range of QDs' size. For this reason, the IHB is assumed about 5 meV. In addition, the HB is assumed 20 meV at room temperature. The parameters of modeled tunneling-injection QD-laser are given in Table 1.
In this paper, the switchable two-color QD-laser using SECs is investigated by assuming the two wavelengths, 1.31 μm and 1.55 μm, which have special positions in telecommunication applications. The radiuses of spherical QDs in the active region are 3.85 nm and 2.47 nm, and the wavelength difference of about 240 nm is an appropriate magnitude to the independent lasing for these wavelengths without overlapping. The prominent idea in this work is introducing the switchable QD-laser, which can provide emission lasing just at one wavelength or at two wavelengths simultaneously. To prove this idea in the following, the output characteristics of this QD-laser is investigated.
Based on described rate equations in the previous section, the output power intensity depends on QD coverage and the photon numbers, which can be controlled by injection current at each wavelength. Thus, the high intensity of output power for two different wavelengths is attained at specific QD coverage, so the coverage of big QDs is considered ξ 1 = 0.07, whereas the small one is ξ 2 = 0.05. Furthermore, at 1.31 μm, the output power value gets efficient due to the effective carrier transition of channel2 (1.31 μm) in comparison with channel1 (1.55 μm) at I th = 14 mA. Also, for both channels, the FWHM of HB and IHB are assumed to 2ℏΓ B = 20 meV and Γ G = 5 meV, respectively. Figure 4(A) illustrates the output power intensity of big QDs ensemble when the current is injected just to the related contact; in other words, there is no current injection to the other contact, so the lasing only occurs at 1.55 μm. The same as the previous performance which was applicable for its contact and is shown in Fig. 4(B), lasing occurs just at 1.31 μm. The super-strength of this suggested model is the selectivity of emission wavelength, in which the existence of different sizes of QDs and the related QW contacts allow controlling the lasing emission wavelength. Figure 4(C) shows the output power when the current is injected into both contacts and the lasing occurs in two wavelengths simultaneously so we can see the output power intensity in both of them, the intensity of output power in channel1 and channel2 equals to 12.3 mW and 12.95 mW. The Fig. 4(D),(E) shows the optical gain spectrum of big QDs (small QDs) when the injection current to channel1 (channel2) is I 1 = 2.5I th = 35 mA.
Since there is no current injection into channel2 (channel1), so the negative gain can be noticed which is determined by the green curve (orange curve). Finally, Fig. 4(F) depicts the optical gain spectrum at both 1.55 μm and 1.31 μm when the current is injected in both channels simultaneously.  Fig. 5(A) by assuming ξ 1 = 0.07 and ξ 2 = 0.05, at injected current I 1 = I 2 = 2.5I th (I th = 14 mA) the transient response of output power characteristics for the switchable two-color QD-laser using SECs are illustrated for both big QDs (orange line) and small QDs (green line) corresponding to the wavelengths of 1.55 μm and 1.31 μm, respectively while the FWHM of HB and IHB are considered to 2ℏΓ B = 20 meV and Γ G = 5 meV. As shown in this figure, the turn-on delay time of big QDs is smaller than the small ones.
The output power as a function of injected currents for switchable two-color QD-laser using SECs has been shown in Fig. 5(B). Here, I i sets to I 1 and I 2 which relates to channel1 (1.55 μm) and channel2 (1.31 μm). The orange solid curve relates to 1.55 μm and the green solid one to 1.31 μm. By increasing the injected current, the output power corresponding to each wavelength is enhanced linearly with different slope efficiency. Here, as it is seen, the threshold current of both channels is almost equal and it is about 14 mA, despite the different considered coverage of big and small QDs, where ξ 1 = 0.07 and ξ 2 = 0.05. In other words, applying injection current in both channels simultaneously where their QD coverage is different, leads to the population inversion occurs at the same current in both big and small QD group. As another impact, the output power variation with injection current has been investigated for two different FWHM of IHB (Γ G = 5 meV and 10 meV). The comparison of  As it is observed in this figure, the QD size distributions (IHB) leads to small variation over the modulation bandwidth which can be ignored (the modulation bandwidths equal about 11 GHz). It should be mentioned that in channel1 if FWHM of IHB takes a high value, the output power declines. So, achieving the efficient operation for channel1 is possible by choosing the suitable FWHM of IHB which can be controlled by ∆R 1 . Consequently, in channel1, the size of QD may have a deviation of ∆R 1 = 0.12 nm for QD radii of R 1 = 3.85 nm which causes the broadening of origin energy. In this case, Γ G = 5 meV and it is worthy of mention that, by using the solution process method, it is possible to obtain very uniform and precise QDs with 0.12 nm size deviation.
Besides, in Fig. 5(D), the solid and dashed green curves are depicted the modulation response behavior of channel2 (1.31 μm). Although increasing the QD size distributions (IHB) can vary the modulation bandwidth, the output power still remains stable. Therefore, the comparison of curves leads to the observation of 11 GHz modulation bandwidth at Γ G = 5 meV while 11 GHz belongs to Γ G = 10 meV. Considering deviation about ∆R 2 = 0.1 nm for a specific small QD radius R 2 = 2.47 nm and the possibility of achieving the unique size by solution process technology, leads to select Γ G = 5 meV for channel2.
The output power intensity corresponding to the wavelengths of 1.55 μm and 1.31 μm as a function of injected current for various QD coverage at both Γ G = 5 meV and 10 meV is demonstrated in Fig. 6(A),(B), respectively. Considering various numbers of coverage for big QDs (channel1 corresponds to 1.55 μm) and small QDs (chan-nel2 corresponds to 1.31 μm) at constant injected current, due to the lack of carriers, the number of occupied big (small) QDs, lying within the scope of central mode of big (small) QDs, decreases when the coverage of big (small) QDs increases; therefore, the corresponding output power intensity reduces, and the threshold current increases as illustrated in Fig. 6(A),(B). In other words, by increasing the coverage of big (small) QDs in www.nature.com/scientificreports www.nature.com/scientificreports/ both channels, higher injection current is needed to achieve the threshold gain and start lasing. As Fig. 6(A),(B) depicts, in all cases by increasing the injection current, the output power increases linearly.
In Fig. 6(A), the current injected in channel1 varies gradually and the current injection to channel2 is constant. Hence, by considering ξ 2 = 0.05, as it is seen, the solid and dashed green lines for both Γ G = 5 meV and 10 meV, shows the output power has constant value equaling to 20 mW. As explained earlier, by comparing the different slop efficiency of output power curves at Γ G = 5 meV and 10 meV, it is revealed that high value of IHB (Γ G = 10 meV) leads to more QD group lying within the HB of central group, while the narrow transmission rate cannot satisfy all the QDs, so at the same QD coverage, the output power intensity of Γ G = 10 meV is lower than it for Γ G = 5 meV.
Similarly, the comparison of curves in Fig. 6(B) demonstrates that the coverage of big QDs is constant (ξ 1 = 0.05) and the current injection into channel1 is constant (solid and dashed orange lines show the output powers at Γ G = 5 meV and 10 meV, which equal to 10.5 mW and 16 mW, respectively). As described previously, increasing the coverage of small QDs causes threshold current increase; also, the slope efficiency of curves for both Γ G = 5 meV and 10 meV isn't different due to the existence of only GS in small QDs and prominent carrier emitting into the central mode.
Based on the transient response result, the modulation response is analyzed for different coverage of big and small QDs for channel1 (1.55 μm) and channel2 (1.31 μm) that are depicted in Fig. 6(C),(D), respectively. As shown in both figures, the modulation bandwidth depends on QD coverage and by their variation, it can change. Therefore, numerical results demonstrate that by increasing QD coverage, a high value of modulation bandwidth can be obtained in both channels. As mentioned above, calculating the modulation behavior is carried out by applying very small step perturbation, which is determined by bias current ((∆I i = 0.05I b ) and depends on threshold current (I b = 2.5I th ).
On the other hand, by increasing QD coverage, higher injection current is needed to achieve the threshold gain, so the threshold current increases. In better words, broader bandwidth can be achieved by high bias current as a result of increasing QD coverage, which leads to a high threshold current. In both Fig. 6(A),(B), the FWHM of HB equals to 20 meV; also Γ G = 5 meV and 10 meV. Both figures illustrate the QD size distributions (FWHM of IHB equal to 5 meV and 10 meV) that lead to small variation over the modulation bandwidth. So, the solution process is a low-cost method which can easily control the QD radii and provide unique QDs with minimum size deviation to achieve high-performance device. The resultant maximum value of modulation bandwidth in channel1 (Fig. 6(A)) at ξ 1 = 0.1 and Γ G = 5 meV is about 14 GHz (solid black curve) and for channel2 ( Fig. 6(B)) at ξ 2 = 0.1 with Γ G = 5 meV is about 18 GHz (solid black curve). Also, it should be borne in mind that both of the output emissions can be modulated independently in this proposed structure.

conclusion
In this paper, the switchable two-color QD-laser utilizing SECs by considering the solution process technology has been proposed for the first time. The realization of controlling the lasing channels at different wavelengths is possible by utilizing the QWs as separate selective energy contacts. Furthermore, it was shown that two-color lasing at 1.31 μm and 1.55 μm was achieved by implementing two different radii of InGaAs/GaAs quantum dots in the active layer. Moreover, simultaneously or individually lasing of the proposed switchable two-color QD-laser is modeled based on modified rate equations. The investigation was carried out by considering inhomogeneous broadening as a result of the size distribution of QDs and considering tunnel injection of carriers into the QDs via SECs. Besides, the effect of QD coverage and changing FWHM of IHB on the threshold current, output power intensity and modulation response of each lasing channel were discussed. Controlling the QD coverage and FWHM of inhomogeneous broadening leads to provide customized output power peak value and modulation bandwidth. Finally, yet important, in this work, the design of switchable two-color QD-laser using separate SECs capability choosing the lasing channel was proposed, and the solution process method as the realization of this structure was presented. Furthermore, both channels can be modulated independently, which is a significant feature in high-speed data transmission.