(53a) Quantum Chemical and in Situ Raman Studies for the Homogeneous Thermal Decomposition Kinetics of Triethylgallium
Triethylgallium ((C2H5)3Ga, TEGa) has been widely used as a precursor in MOCVD of Ga-V compound semiconductors and their alloys. Although its synthesis was reported by Dennis and Patnode1 almost 80 years ago, there have been few reports on the thermal decomposition pathways important to MOCVD. Experimental studies of the homogeneous thermal decomposition of TEGa have suggested both homolysis of ethyl groups and b–hydride elimination pathways.2-4 The reported experimental activation energy for removal of the first ethyl group, however, ranges from 22.6 to 47.2 kcal/mol, and likely related in part do use of different carrier gas, i.e. toluene or hydrogen.2, 5-6 In this study, TEGa decomposition kinetics have been examined in a vertical upflow, cold-wall CVD reactor by in situ gas-phase Raman spectroscopy (Figure 1). The CVD reactor can be translated x-y-z to yield temperature profile by analysis of the carrier gas rotational bands and composition profiles by analysis of the vibrational bands. These profiles were then compared with those using model calculations (FEM) of the reactor.
Figure 1. Schematic of the experimental reactor for in situ Raman spectroscopic measurements
In a preliminary experiment, the relative Raman cross-section of TEGa was measured at room temperature and atmospheric pressure. In this experiment, a 2.5 cm/s steady flow of 5 mol % TEGa in N2 carrier gas was injected along the reactor center line, while annular and sweep flows of pure N2 were introduced with matched velocity of 2.5 cm/s. The 1.5W Nd:YAG solid-state laser line was used to excite the TEGa source and the [Ga–C3] vibrational Raman excitation line (490cm-1 line) was recorded (Figure 2).
Figure 2. Recorded Raman intensity of the [Ga–C3] skeleton vibration (490 cm-1).
Although the scattered intensity of TEGa is much weaker than many other group II and III alkyls (e.g., TMGa: 17.507, TMIn: 22.38), a reasonable signal to noise ratio was obtained with repeated measurements and long integration time at room temperature. The relative Raman cross-section of TEGa was estimated to be 2.7.
In the next set of studies, 5 mol% TEGa was again introduced along the reactor centerline in N2 carrier, impinging on a resistively heated susceptor with set point 750 °ÆC. The relative mole fraction of each gas-phase chemical species along the reactor centerline was obtained from the ratio of integral of the primary peak [Ga–C3] at 490 cm-1 to that of the N–N vibration at 2331 cm-1. Using the custom-designed FEM simulation, which has been described in detail elsewhere9, 10, we can trace TEGa concentration based on various mechanism assumptions as shown in the Figure 3. From the simulation results, the rate of disappearance of TEGa cannot be explained with a single reaction, suggesting that both b-hydride elimination and homolysis take place simultaneously under these conditions. Moreover, experimental concentration profile agrees well with the results that include both reactions. In addition, the case for no reaction is also shown in the Figure 3 along with the predicted gas-phase temperature profile, which compared well to the measured profile.
Figure 3. Simulated and experimental concentration profile of TEGa as a function of distance from the heated susceptor (750 °).
Since both b–hydride elimination and Ga-C bond homolysis are have been suggested in the literature and by the analysis of the scattering data of this study, quantum chemical calculations were performed to better understand both pathways. These calculations used the Gaussian 03 suite11 to represent species in the TEGa decomposition system. Based on TEGa structure data, the B3LYP level calculation with LanL2DZ basis set was chosen. The Berny algorithm was employed for geometry optimization. For the initial stages of both reactions, NBO (Natural Bond Orbital) analysis was employed to suggest the weakest bond and reaction directions. For the homolysis pathway, since the Wiberg bond indices between Ga and Ca (0.5976) are much smaller than that between Ca and Cb (1.0563), Ca–Cb breakage in the ethyl group was not considered . The b–hydride elimination case is more complicated since a suitable transition state must be envisioned that includes the imaginary bond between hydrogen connected with Cb and the gallium center. Additional information on the imaginary ring consisting of gallium, Ca, Cb and hydrogen connected to Cb was analyzed from Wiberg indices change. Since the energy for the internal rotation of one ethyl group is small (~2 kcal/mol), the rotational conformer that is necessary for the b–hydride elimination can be constructed easily in the CVD reactor. The theoretical enthalpy (¥ÄH), entropy (¥ÄS) and Gibbs energy (¥ÄG) of various association and dissociation reactions including homolysis and b–hydride elimination reactions in the thermal decomposition pathways of TEGa at 298K and 900K as well as the possible reaction pathways leading to formation of the 4-membered-ring, i.e. (MEGa)4, consisted of 35 species and 39 reactions including three transition states of b–hydride elimination. More detailed computational calculations were performed for the two initial stages of TEGa thermal decomposition using same model chemistry as shown in Figure 4.
Figure 4. Calculated energetics of the two major thermal decomposition pathways of TEGa with reaction enthalpies at 298K (kcal/mol)
Quantum chemical calculation results of this study suggest a relatively high 65.01 kcal/mol for homolysis and 40.30 kcal/mol for b–hydride elimination as activation energies and 52.23 and 30.63 as frequency factors in natural logarithm scale, respectively. These values are calculated from the Arrhenius plot in the temperature range 400~900K. Calculated results of this study show good agreement with reported values.
The results of vibrational frequency calculations were used to describe the decomposition behavior and assign the observed Raman bands (490cm-1, 517cm-1, 537cm-1, and 555cm-1) between gallium and a–carbon to the decomposition products (Et)3Ga, (DEGa)2, (Et)GaH–Ga(Et)2 and (Et)GaH–GaH2, respectively. In addition, reasonable transition states during b–hydride elimination were identified, when varying the directionality of ethyl group, i.e. propeller-like and confronted arrangements. The experimental and computational results of this study are consistent with the coexistence of both b–hydride elimination and homolysis reactions. The measured composition and temperature profiles were used to estimate rate constants using a 2-D detailed reactor model. DFT calculations using B3LYP/LanL2DZ model chemistry were performed to screen possible decomposition routes and estimate their reaction rate parameters.
1. Dennis LM, Patnode W, Gallium triethyl etherate, gallium triethyl, gallium triethyl ammine, J. Am.Chem. Soc. 1932; 54: 182-188.
2. Paputa MC, Price SJW, Pyrolysis of triethylgallium by the toluene carrier technique, Can. J. Chem.1979; 57: 3178-3181.
3. Mahmood Z, Hussain I, Linney RE, Russel DK, Comparative infrared laser-powered homogeneous pyrolysis studies of triethylgallane, trimethylgallane, triisopropylgallane, triisobutylgallane, and tri-tert-butylgallane, J. Anal. Appl. Pyrolysis 1997; 44: 29-48.
4. Russell DK, Mills GP, Raynor JB, Workman AD, Radical Processes in the Pyrolysis of TrialkylCompounds of Group 13 CVD Precursors, Chem. Vap. Deposition 1998; 4: 61-67.
5. Yablokov VA, Yablokova NV, Kinetics of the thermal decomposition of alkyl derivatives of GroupIII and V elements, Russ. Chem. Rev.1995; 64: 953-966.
6. Grady AS, Linney RE, Markwellt RD, Russell DK, Exchange and Pyrolysis in Mixed Trialkylgallanes, J. Mater. Chem.1993; 3: 483-488.
7. Huang M. Ph. D. Thesis, University of Florida; 2003.
8. Hwang JY. Ph. D. Thesis, University of Florida; 2004.
9. Hwang JY, Park CH, Huang M and Anderson TJ, Investigation of mass transport phenomena in anupflow cold-wall CVD reactor by gas phase Raman spectroscopy and modeling, J. Cryst.Growth, 2005; 279: 521-530.
10. Park CH, Hwang JY, Huang M and Anderson TJ, Investigation of an upflow cold-wall CVD reactor by gas phase Raman spectroscopy, Thin Solid Films, 2002; 409: 88-97.
11. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Montgomery JJA, Vreven T, Kudin KN, Burant JC, Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B, Cossi M, Scalmani G, Rega N, Petersson GA, Nakatsuji H, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Klene M, Li X, Knox JE, Hratchian HP, Cross JB, Bakken V, Adamo C, Jaramillo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C, Ochterski JW, Ayala PY, Morokuma K, Voth GA, Salvador P, Dannenberg JJ, Zakrzewski VG, Dapprich S, Daniels AD, Strain MC, Farkas O, Malick DK, Rabuck AD, Raghavachari K, Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clifford S, Cioslowski J, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A, Challacombe M, Gill PMW, Johnson B, Chen W, Wong MW, Gonzalez C, Pople JA, Gaussian 03, Revision B.04, 2004.