Inkjet 3D printed chip for capillary gel electrophoresis

Highlights

• We presented for the first time inkjet 3D printed chip for on-chip gel electrophoresis.

• We found that printing orientation of the microchannels plays a crucial role in the proper operation of the separation process.

• The building material used in this study is semi-transparent and autofluorescence of the material is not noticed, thus fluorescence detection is possible.

• 50–800 bp DNA ladder can be separated in the printed chip with fluorescence detection at the end of the separation microchannel.

Abstract

This paper presents for the first time the use of an inkjet 3D printing to develop a chip for capillary gel electrophoresis. The designing of the chip is preceded by investigations into surface roughness and the geometrical properties of the printed microchannels, which are important considerations for the separation process. The optical properties of the building material are also determined to confirm whether fluorescence detection is possible. It is found that the printing orientation of the microchannels plays a crucial role in the proper operation of the separation process. A description of the successful separation of a 50–800 bp DNA ladder, which was achieved with fluorescence detection at the end of separation microchannel, is given. Based on electropherograms, the number of theoretical separation plates is calculated (max. 70,000) and compared with data from the literature. Finally, some conclusions and observations on the design of the chip are given in order to improve its configuration in the future toward modular configuration.

Introduction

On-chip capillary gel electrophoresis is one of the main techniques used to analyse genetic material. Numerous scientific papers and application reviews have been published on this technique in recent years [[1], [2], [3], [4], [5], [6], [7], [8], [9]]. On-chip gel electrophoresis is an alternative to the widely used standard slab gel electrophoresis, and it was one of the first techniques to demonstrate the advantages of scaling down conventional instrumental techniques to a microfluidic format. By decreasing the length of the separation column from several decimeters to a few centimeters or less, the duration of a single analysis is significantly reduced, much lower voltages can be used, and sample plug injection and detection can be integrated. Generally, it enables fast and accurate separation and detection of genetic material through the use of disposable chips. Our recent works on on-chip electrophoresis with sieve matrix have demonstrated that it is a very powerful technique, and can reliably detect the avian influenza virus within minutes when chip construction, separation conditions and optical detection methodology are optimized [10,11].

The chips for on-chip gel electrophoresis are made of many materials commonly applied in lab-on-a-chip technique i.e. glass, silicon, PDMS and other polymers [[1], [2], [3], [4], [5], [6], [7], [8], [9], [10]] with glass as the main one. Various kinds of glass (borosilicate, soda, quartz) are used due to their well-established (but not easy) microengineering and the great chemical, physical and optical properties of glass [10].

Usually, the manufacturing of a chip involves the processing of at least two substrates. The opened microfluidic channels that determine the dosage amount (if necessary) and the separation channels are fabricated in the first substrate using one subtractive technology (e.g. etching or milling) or injection/embossing. Next, the second substrate is processed to form inlet/outlet holes. In some technical realizations, reservoirs for buffers are also fabricated and bonded to the second substrate. Finally, all substrates are bonded together to form tightly closed microfluidic channels. This multistep and time-consuming fabrication of the chip involving expensive equipment and reagents is necessary to manufacture a mainly disposable chip. For an all-glass chip with integrated buffer reservoirs, the estimated cost of the borosilicate glass material is around 40 Euro per chip and the fabrication time, involving HF-based etching of microchannels, drilling the holes, cutting and polishing the reservoirs, cleaning the surfaces and finally thermally bonding all the components, is 36 h (based on our experience).

On the other hand, three-dimensional printing (3DP) of microfluidic structures is a very attractive alternative to typical techniques used for lab-on-a-chip fabrication. Although 3DP is not a new technique (stereolithography was introduced in the mid-1980s), rapid growth in the number of scientific papers describing the applicability of 3DP in microfluidics has been observed in recent years [12]. This is because the technique results in easy fabrication, can provide on-demand customization and personalization of even complex structures, and creates less waste in comparison to other techniques.

Common 3DP techniques used in lab-on-a-chip fabrication are stereolithography (STL), inkjet printing (i3DP) and fused deposition modelling (FDM). Every several months, comprehensive reviews on new applications of these techniques are presented in well-recognized scientific journals [[12], [13], [14], [15], [16], [17], [18], [19]]. These reviews, as well as other regular papers on the subject, show how 3DP has been adapted to fulfill the requirements of microfluidics and then applied in various applications. 3DP was first characterized as a potential tool for fabricating microfluidic structures, then simple passive microfluidic structures (i.e. microchannels, micromixers, etc.) were fabricated and analysed. Some limits of STL, i3DP and FDM were recognized in this analysis, i.e. printing resolution, surface roughness, support material removal and biocompatibility problems [[20], [21], [22], [23], [24], [25]]. In the meantime, 3DP molds were developed for fabricating PDMS devices [26,27] and more complex microfluidic structures for chemical and biochemical analysis were fabricated using 3DP and reviewed [28]. Finally, active components like valves and pumps were developed and successfully applied in microfluidic platforms [29,30].

To the best our knowledge, gel electrophoretic separation has not yet been analysed in 3DP microfluidic chips. Anciaux et al. presented a 3D printed micro free-flow electrophoretic device fabricated using FDM [31], but the chip was fabricated from two separately printed substrates and then bonded together. Thus, one of the main advantages of 3DP – the generation of a complete microfluidic structure in a single printing process – was lost. Cabot et al. described a fiber-based electrofluidic device utilizing a low cost versatile 3D printed platform for solute delivery, separation and diagnostics [32], but the 3D printed elements (the platform) formed only part of the device based on commercially available separation fibers.

We suppose that the lack of reports on the applicability of 3D printed microfluidic chips for on-chip gel electrophoresis is due to two main factors. Firstly, in order to conduct on-chip gel electrophoresis, it is necessary to fabricate relatively long (tenths of millimeters) separation channels with small diameters (up to hundreds of micrometers) to ensure proper separation efficiency when filled with a sieve matrix. Secondly, there is no information about the optical properties of the applied printing material from the point of view of the most commonly applied detection method for on-chip gel electrophoresis – fluorescence.

In this paper, we present for the first time the successful application of i3DP for fabricating a microfluidic chip for gel electrophoresis. We find the optimal printing conditions and support material removal procedure to enable the fabrication of short dosing microchannels and long separation microchannels. The optical properties of the building material (i.e. visible light (VIS) transparency and autofluorescence), which are important for fluorescence detection, are determined to find the optimal fluorescence excitation/detection wavelength region. Finally, the chip is designed, fabricated and tested with a sieve matrix and DNA ladder to show the usefulness of the developed construction as well as its limitations.