半导体外文翻译--半导体制造过程控制和监测：工厂全框架(编辑修改稿)

半导体外文翻译--半导体制造过程控制和监测：工厂全框架(编辑修改稿)内容摘要：

2,41]. Research groups at University of Maryland contributed in the area of run to run control [2,5,53]. SEMATECH, a consortium of leading semiconductor manufacturers, posted several benchmark problems on plasma equipment fault detection and diagnosis [4]. Adaptive and nonlinear control for R2R operations is proposed by [16]. Model predictive control is applied to R2R control as well which has additional capability in handling constraints explicitly [19]. At UTAustin we have developed (i) stability conditions and tuning guidelines for multivariable EWMA and double EWMA control with metrology delays [20,21], (ii) multivariate statistical monitoring of RTA and etchers [52,51], and (iii) multivariate statistical control of CD metrology data from lithography [14]. Other new development and applications of control and fault detection are reported at recent SPIE conferences and AEC/APC Symposia organized by SEMATECH and summarized in Del Castillo and Hurwitz [15] and Moyne et al. [34]. Manufacturing panies like AMD, Intel, Motorola, and TI and vendors like Applied Materials, BrooksPRI Automation, and Yield Dynamics are leaders in deploying APC technologies at the manufacturing lines. In this paper we draw the analogy between semiconductor manufacturing fabs and chemical plants and propose a hierarchical optimization and control system for semiconductor fab control. A schematic diagram is shown in Fig. 1 for this analogy, which was first presented by Qin and Sonderman [38]. The equipment level control involves automatic feedback control of tool parameters and small scale runtorun control using integrated metrology. The next level runtorun control involves the use of inline measurement for feedforward and feedback control. The third level is the islands of control. The top level of the hierarchy is the fabwide control which is the highest level optimization to achieve desired electrical properties by recalculating the optimal geometric targets and dosage for the lower level. The organization of the paper is given as follows. We first propose a hierarchical fabwide control strategy with the integration of 300 mm equipment and metrology tools and highly automated material handling system. Relevant runtorun technology is reviewed and analyzed in the fabwide control context, process and metrology data monitoring are discussed with an example, and missing ponents are pointed out as opportunities for future research and development. Concluding remarks are given at the end of the paper. 2. A framework for fabwide control Almost all existing development is on R2R control which adjusts recipes of a step based on metrology data at the equipment level. These are known as islands of control as illustrated in the lower part of Fig. 1. None of the existing control strategies examine the coordination of multiple manufacturing steps to improve the overall product quality in terms of electrical parameters. The R2R controllers pensate for equipment drifts through metrology feedback, but they cannot pensate for metrology drifts and uncertainties. The direct control of electrical parameters proposed here can pensate for metrology drifts and systematic errors in the geometric measurements that are below the metrology SPC limits. It is believed that the control and optimization of electrical parameters represent the next generation of semiconductor manufacturing control system as it directly controls the electrical properties to a desired product profile by manipulating the operation requirements for lower level R2R controllers. The electrical parametric control and optimization will maximize the yield of highgrade products or reduce operational cost when a demand profile is specified by market orders. The fabwide control framework in Fig. 1 provides optimization and coordination from step to step to reduce variability, reworks, and scraps, thus improving the overall equipment effectiveness and reducing manufacturing cost. This framework was first presented by Qin and Sonderman [38] after having deployed many R2R controllers at AMD and analyzed the need for a higher level control. The equipment level control involves automatic feedback control of tool parameters. The next level is runtorun control using integrated or inline metrology to achieve a specified target. The third level is the islands of control that shares information from multiple steps to perform feedforward and feedback control and tool performance matching. The top level of the hierarchy is electrical parametric control (EPC) or fabwide control to achieve desired electrical properties by recalculating the optimal targets for the lower levels. Equipment drifts, metrology drifts, and material variations are pensated by feedback at the EPC level, leading to improved process and metrology availability and reduced use of calibration and test wafers. This multiple level control framework resembles the hierarchical control framework that has been successful in the refinery industry [39], but significant differences exist: (i) the lowest level control is mostly batch operations。 (ii) the middle level R2R control has virtually no R2R process dynamics except for disturbance dynamics。 and (iii) the top level EPC is a multistep operation control that aims to pensate for errors made in prior steps, regardless of the nature of the errors as long as stepwise metrology measurement is available. This makes it different from model predictive control (MPC) of batch processes with shrinking horizons. In chemical and refinery process, the toplevel optimization is realtime optimization [31] and the middlelevel is the fullscale dynamic MPC. As the MPC framework is a powerful and successful technology, it has been extended to scheduling and production planning in the semiconductor industry [11,46,47].。