Conventional photolithography and photoresist stripping processes were employed to construct channels with

the desired depth. A silicon (Si) wafer was cleaned in H2SO4:H2O2 solution find protocol (volume ratio of 10:1) at 120°C for 10 min, followed by deionized water (DI) for 4 cycles, then HF:H2O solution (1:50) at 22°C for 1 min and DI water for 4 cycles, and finally spin-dried in hot N2 gas for 15 min. Then, the Si wafer was processed by hexamethyldisiloxane (HMDS) coating and positive photoresist HPR 504 spin-coated at 4,000 rpm for 30 s. The wafer was soft-baked on a hot plate at 110°C for 60 s before exposing to UV via the Mask Aligner (SUSS Microtec MA6-2, Garching Germany) for 5 s. The photoresist was developed using FHD-5 for 60 s and post-baked on a hot plate at 120°C for 60 s. The micropatterns were successfully defined at this stage. The Si wafer was then selleck etched by a DRIE machine (Surface Technology Systems, Newport, UK) and followed by photoresist stripping in PS210 Photoresist Asher (PVA Tepla AG, Kirchheim, Germany) for 25 min. After constructing the microchannels, 10 nm of thermal oxide was grown using a diffusion furnace to form silica on the channel wall. After drilling the inlets and outlets on the Si chip by a mechanical driller, the chip has to be sealed to form a closed channel. A thin film of polydimethylsiloxane (PDMS) was applied for such purpose due to the good adhesion between PDMS and the Si chip. PDMS was formulated

from Sylgard 184 silicone elastomer mixture (Dow Corning Corporation, Midland, MI, USA) at a weight ratio of base:curing agent = 10:1. Then, it was poured onto a Si wafer with saline coating on the surface and pressed against a cleaned glass slide. After curing PDMS in an oven at 60°C for 2 h, the microchip was constructed by pressing the Si chip against the glass slide Vildagliptin with the thin layer of PDMS on its surface. The fabricated microchip is shown in Figure  2a. The microreactor is comprised of two microchannels: channels A and B with a width of 300 μm and a depth of 12 μm and an array (20 channels) of 1D nanochannels that connected the two microchannels

to demonstrate the injection process. It is not necessary to adopt 20 nanochannels. One can increase or decrease the number according to their applications. Fewer nanochannels will result in higher precision, and more nanochannels will give a higher throughput. The inset (a1) in Figure  2a illustrates the multilayer structure showing the PDMS, the silicon chip, and the glass slide. Another inset (a2) shows the structure of the two microchannels connected by the nanochannel array that is highlighted by the green dashed square. When the electric field across channel A and channel B was applied, fluid flowed from channel A to channel B through the nanochannel array as indicated by the green arrow in the same figure. The enlarged scanning electron microscopy (SEM) image of the nanochannel array is shown in Figure  2b. The channel width observed was 10 μm.