论文部分内容阅读
砷和硼从预淀积层(化学源或离子注入)作同时扩散,已被用来分别制作微波晶体管的发射区和基区。这类晶体管掺杂分布的数学模拟表明:在预淀积——扩散结构中,砷-硼(化学源)的相继扩散中所发生的扩散互作用效应是不显著的。其结果是,由这样的工艺制造的晶体管没有基区迟滞现象,而且其有源基区掺杂浓度要比由以前确立的扩散方程所预计的高些。为了确定互作用效应重要到何种程度,测量了晶体管的掺杂分布,并与计算的分布作了比较。把电场的相互作用、[V_(si)As_2]络合物形成而引起的空位欠饱和的状况以及离子配偶都包括在内,估计出每一种作用对于硼扩散的重要性是可能的。业已表明,在预淀积——扩散结构中,其电场互作用要比在恒定表面浓度扩散(非耗尽源)中的小2到3倍。更重要的是,在从预淀积层同时扩散砷-硼期间,发生的空位欠饱和是可忽略的。在化学源砷预淀积的情况中,这是由于在预淀积期间(在与硼同时扩散之前)就达到了一个[V_(si)As_2]络合物的准平衡浓度。在离子注入预淀积砷的情况中,对于剂量≤3×10~(15)厘米~(-2)时,络合物好像是或者在注入期间形成的,或者在退火期间非常迅速地形成的。给出的数据表明,在砷注入——退火结构中,没有非电活性的砷络合物形成,在这个结构中,已接近As~+离子的最大溶解度了(在1000℃下或剂量为5～8×10~(15)厘米~(-2)时为3.8×10~(20)厘米~(-3))。这个结果适合于本研究中所用的砷剂量。由于这个结果在由化学源作扩散的砷掺杂层中没有观察到,为了解释这个异常现象,尚需进一步作研究工作。
The simultaneous diffusion of arsenic and boron from a pre-deposited layer (chemical source or ion implantation) has been used to create the emitter and base regions of a microwave transistor, respectively. The mathematical simulation of the doping profile of such transistors shows that the diffusion-interaction effect that occurs in the sequential diffusion of arsenic-boron (chemical source) is insignificant in the pre-deposition-diffusion structure. As a result, transistors fabricated in such a process have no base-region hysteresis and their active base doping concentration is higher than expected from previously established diffusion equations. In order to determine the degree of importance of the interaction effect, the doping profile of the transistor was measured and compared with the calculated profile. It is possible to estimate the importance of each of these effects for boron diffusion by including the electric field interaction, the state of under-saturation due to the formation of the [V_ (si) As_2] complex, and the ionosphere. It has been shown that in pre-deposition-diffusion structures, the electric field interaction is 2 to 3 times smaller than in constant surface concentration diffusion (non-depletion source). More importantly, under-saturation of vacancies occurring during simultaneous diffusion of arsenic-boron from pre-deposited layers is negligible. In the case of chemical source arsenic pre-deposition, this is due to the quasi-equilibrium concentration of a [V_ (Si) As_2] complex reached during pre-deposition (prior to simultaneous diffusion with boron). In the case of ion-implanted pre-deposited arsenic, the complex appears to be formed either during implantation or very rapidly for doses <3 x 10 ~ (15) cm ~ (-2) . The data presented shows that there is no formation of a non-electroactive arsenic complex in the arsenic implanted-annealed structure in which the maximum solubility close to the As ~ + ion is reached (at 1000 ° C or a dose of 5 ~8 × 10 ~ (15) cm ~ (-2), the value of 3.8 × 10 ~ (20) cm ~ (-3)). This result is suitable for the arsenic dose used in this study. Since this result was not observed in arsenic-doped layers diffused by chemical sources, further work is needed to explain this anomaly.