The achievement of a good high-k oxide/InGaAs interface quality is a key challenge to obtain high performance MOSFET. Associating Al2O3 and HfO2 in a bilayer oxide is interesting to benefit from the good interface quality obtained with Al2O3 and the better electrostatic control achievable with HfO2. Furthermore, some recent work evidenced the reduction of the interface trap density (Dit) by using a plasma nitridation process [1]. For the first time, we evaluated the passivation properties of a Al2O3/HfO2 bilayer and a nitridation treatment on industrial equipment compatible with 300 mm Si wafers. First, we investigated the number of Al2O3 ALD cycles required to obtain a good interface. Second, we evaluated the effect of a nitridation treatment on the Al2O3/InGaAs interface properties performed on a 300 mm capacitive plasma tool.MOSCAP structures were fabricated on a 27 nm thick In0.53Ga0.47As layer grown on InP substrates. The samples were cleaned in a NH4OH solution (4%) for 1 min at room temperature and rinsed in deionized water. Then, Al2O3 and HfO2 films were deposited in a ALD chamber at 300°C with trimethylaluminium (TMA), Hafnium tetrachloride (HfCl4) and H2O as precursors. The ALD cycle numbers for Al2O3 were 0, 3, 5, 8 and 10 and 32 for HfO2. A post-deposition annealing was carried out at 370°C for 30 min in N2 ambient. For the nitridation study, one InGaAs sample was directly treated with a NH3 plasma for 120 s at 50W and two samples were treated with a NH3 plasma or a N2 plasma for 120 s at 50W after the deposition of Al2O3 (10 ALD cycles) to prevent damages on the III-V layer. After the plasma treatment, Al2O3 was again deposited to reach a 8 nm-thick layer for electrical characterization. Ni/Au gate electrode was deposited through a shadow mask by e-beam evaporation.The sample without Al2O3 (Fig1.a) presents distorted profiles, especially at 10 kHz, sign of a poor InGaAs/HfO2 interface quality. Depositing three Al2O3 cycles (fig.1.b) allows a limited improvement of the characteristics. The dispersion in accumulation of the 10 and 30 kHz curves is reduced but no clear accumulation plateau can be seen. From 5 to 10 cycles, samples (c-e), the situation is significantly improved with a clear accumulation regime. The interface trap densities (Dit) were estimated using the conductance method [2] and were found equal to ~11, 6.1 and 5.5 ×1012 cm-2eV-1 for the 5, 8 and 10 cycles samples respectively. A clear improvement is obtained from 5 to 8 cycles while only a slight improvement is obtained from 8 to 10 cycles. Compare to the Al2O3/InGaAs capacitor (fig.1.f), the best compromise between low Dit and high level capacitance appears to be 8 Al2O3 cycles. Wide frequency dispersion in accumulation can be seen after NH3 treatment without the Al2O3 protection (fig.2c) due to border traps in the dielectric. The plasma treatment prior to the dielectric deposition induces defects at the beginning of the nucleation. This dispersion is reduced when the treatment is implemented through an Al2O3 layer (fig.2b and d). N2 plasma reduces the capacitance whereas NH3 plasma does not deteriorate the C-V characteristics and keeps the same capacitance as the sample without nitridation. According to XPS analysis, N is clearly detected after plasma treatment. The N1s peak can be divided into two components, N1s A at 399.0 eV and N1s B at 397.9 eV. The N1s A component has been attributed to N-O bonds, and N1s B to N-Ga bonds [3]. Residual As oxides are reduced with NH3 (fig. 3e and f) but no As nitride is evidenced. After plasma treatment, the Ga3+ peak increases significantly (fig.3c and d), because of N incorporation at the interface. The Ga3+ component can be attributed to Ga-O and Ga-N bonds resulting in an oxynitride formation [1].In conclusion, Al2O3/HfO2 bilayer was optimized on a 300 mm equipment to achieve low Dit and high level capacitance (1.75 µF/cm²). NH3 nitridation process performed on a 300 mm capacitive plasma tool integrates nitrogen at the InGaAs surface without deterioration of the C-V characteristics. Low Dit is estimated at 2×1012 cm-2eV-1 at room temperature after NH3 plasma which is close to the value shown in [1] obtained with an ECR plasma. Measurements at lower temperature will further explain the Ditenergy distribution. [1] T. Hoshii, J. Appl. Phys., vol. 112, no. 7, p. 073702, 2012.[2] R. Engel-Herbert, Appl. Phys. Lett., vol. 97, no. 6, p. 062905, 2010.[3] T. S. Lay, J. Vac. Sci. Technol. B Microelectron. Nanom. Struct., vol. 22, no. 3, p. 1491, 2004. Figure 1
Read full abstract