RIMS (real-time imprint monitoring by scattering of light) study of pressure, temperature and resist effects on nanoimprint lithography

To optimize nanoimprint lithography (NIL), it is essential to be able to characterize and control the NIL process in situ and in real time. Recently we have developed a real-time imprint monitoring by the scattering-of-light (RIMS) approach, which allows us to detect the degree of resist deformation and the duration of resist penetration by a mould during the imprint process in real time. In this paper we report the performances of RIMS under a broad range of working conditions. RIMS data shows that the resist penetration is facilitated by increasing processing temperature, pressure and the resist film thickness; a prolonged pre-NIL resist baking step, on the other hand, has the effect of slowing it down. Our results provide further demonstration of the effectiveness of this method under different working conditions. RIMS measurements show not only how long an imprint takes to complete, but also how an imprint progresses with time and how it is affected by differences in processing parameters. These measurements provide information crucial for a better understanding and process optimization in NIL. (Some figures in this article are in colour only in the electronic version) Nanoimprint lithography (NIL) is a high throughput, lowcost lithography tool with the potential to become a mass manufacturing process for nanoscale devices and systems [1]. It has been investigated vigorously and has recently demonstrated a 5 nm linewidth resolution at 7 nm halfpitch [1–4]. Unlike other conventional lithographies, polymer deformation plays a critical role in nanoimprint lithography. Instead of using radiation, NIL patterns by physically deforming a resist thin film using a mould with threedimensional topography to form a negative replica, which can be done either by embossing a thermoplastic polymer at an elevated temperature (thermal-NIL) or by pressing a lowviscosity photocurable resist at room temperature and curing it with UV exposure (UV-NIL) [4, 5]. Clearly, a good understanding of the polymer flow during NIL is essential for the rational design of NIL tools and the optimization of NIL processes. Specifically, questions such as how is the speed of resist penetration affected by differences in processing parameters (temperature, pressure, type and thickness of the resist, pre-NIL baking conditions, etc) need to be answered. Previously, effects of resist thickness, resist viscosity and mould geometry have been studied by simulations or indirect measurements. The usefulness of this kind of investigation is limited because it does not provide the crucial time-dependent information on the transient state in the NIL process [6, 7]. Without an effective means for in situ NIL process detection, to ensure a complete deformation of the resist, it was a common practice to hold the mould and resist at a high processing temperature (in some cases ∼100 ◦C above Tg) under a high pressure for a long duration of time. Obviously, these processing parameters should be optimized: on the one hand, high temperature and pressure may not be 0957-4484/07/065304+04$30.00 1 © 2007 IOP Publishing Ltd Printed in the UK Nanotechnology 18 (2007) 065304 Z Yu et al pm wm hm hp mold (nm) substrate resist (nr) P (a) (b)