*3.2.1 Structures filling mechanism for micro injection molding*

**Figure 10(a)** illustrates a fast prototyping mold for the injection molding of a flow cytometer chip, which has an area of 26.12 mm × 26.12 mm and a thickness of 1.05 mm [50]. **Figure 10(b)** shows surface structures that are used to fabricate inverted channels, all of which are 250 μm wide and 150 μm deep for the parts that had rectangular cross-sectional areas. The horizontal channel works as a fluidic channel, in which biological particles, such as cells, are focused by sheath flow from side inlets. Meanwhile, an optical fiber is integrated into the microfluidic chip using tilted channels. One channel is used for a fiber laser excitation beam and the other two are allocated for detection of forwarding scattering and side/fluorescence scattering by fibers. In order to minimize light intensity reduction of the excitation

#### **Figure 10.**

*Flow cytometer chip: (a) cassette mold, (b) three-dimensional model of surface structures on the chip, (c) actual chip assembly, (d) connection with tubing [55].*

**43**

**Figure 11.**

*Prototyping and Production of Polymeric Microfluidic Chip*

laser and to decrease potential scattering when the laser beam passes through plastic material, the distance between the fluidic channel and optical fiber channels has to be as small as possible. This requirement necessitates surface structures as shown in the upper image in **Figure 10(b)**. The protruding structures demonstrate a gradual aspect ratio, and the highest ones are at the places between the edge and the fluid channel, which have an aspect ratio of 3:1 nominally; it is these that were mainly investigated. **Figure 10(c)** shows the actual microfluidic chip (after bonding), which is assembled into a designed chip holder and connected with samples and PBS (Phosphate-buffered saline) sheath flow using PTFE (Polytetrafluoroethylene)

In theory, the surface structures should fill more easily under such a layout because the gate was parallel to the feature of interest and the centre fluid channel [51]. However, replication of micro/nanoscale surface structures is challenging due to their high surface to volume ratios, especially for high aspect ratio features. Fast heat transfer rates mean that the polymer melts are inclined to solidify before the cavities are fully filled. For example, experimental results show that the cross-sectional area of the surface structure emphasized in **Figure 10(b)** only reaches up to 70.67% of the criterion. Combined with microscopy, process monitoring, and morphology, A combined melt flow and creep deformation model is proposed to explain the complex filling behavior of the surface structure. As shown in **Figure 11**, the overall replication is ordinarily composed of two parts: melt flow during the injection stage and creep deformation during the packing stage. Melt flow is related to injection velocity and pressure, while the extent of creep deformation is linked to mold temperature, packing pressure, packing time, etc. Based on this, increasing shot size to improve the replication quality of surface structures is proposed, and the replication is significantly improved and sufficient to satisfy the practical requirements of a

In order to better understand the connection between product quality and process parameters, process simulation of injection molding has developed over several decades. Simulation in the early stages of part and mold design is relatively cost-efficient and offers the capability to evaluate various design options, such as runner design and gate designs. However, microscale effects, such as altering heat transfer coefficient (HTC), wall slip behavior, mold surface roughness, venting operations, which tend to be ignored in conventional injection molding simulation, should be considered in the simulation of microinjection molding effectively [52–54]. Another important concern is that most studies only used nominal machine processes, but they do not include the actual machine dynamics, which is important for microinjection molding of surface structures because microscale surface structures are much sensitive to process variation. As a result, simulation results are more or less unconvincing and cannot be adapted for real-world applica-

*Filling mechanism of surface structure on a substrate: (a) the injection stage, (b) the end of the injection stage,* 

tions, and the simulation inaccuracy needs to be addressed.

*(c) the packing stage [50]. Copyright (2018) with permission from IOP publishing, LTD.*

*DOI: http://dx.doi.org/10.5772/intechopen.96355*

tubing (**Figure 10(d)**).

microfluidic flow cytometer chip.

*3.2.2 Process simulation and validation*
