High-throughput controllable generation of droplet arrays with low consumption

High-throughput controllable generation of droplet arrays with low consumption

Accepted Manuscript Full Length Article High-throughput controllable generation of droplet arrays with low consumption Yinyin Lin, Zhongsheng Wu, Yibo...

939KB Sizes 0 Downloads 29 Views

Accepted Manuscript Full Length Article High-throughput controllable generation of droplet arrays with low consumption Yinyin Lin, Zhongsheng Wu, Yibo Gao, Jinbo Wu, Weijia Wen PII: DOI: Reference:

S0169-4332(18)30398-2 https://doi.org/10.1016/j.apsusc.2018.02.055 APSUSC 38513

To appear in:

Applied Surface Science

Received Date: Revised Date: Accepted Date:

19 December 2017 1 February 2018 5 February 2018

Please cite this article as: Y. Lin, Z. Wu, Y. Gao, J. Wu, W. Wen, High-throughput controllable generation of droplet arrays with low consumption, Applied Surface Science (2018), doi: https://doi.org/10.1016/j.apsusc.2018.02.055

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

High-throughput controllable generation of droplet arrays with low consumption Yinyin Lina, Zhongsheng Wua, Yibo Gaob, Jinbo Wua* and Weijia Wenc a

Materials Genome Institute, Shanghai University, Shanghai 200444, China

b

Shenzhen Shineway Technology Corporation, Shenzhen, China

c

Department of Physics, The Hong Kong University of Science and technology, Hong Kong, China

We describe a controllable sliding method for fabricating millions of isolated femto- to nanoliter-sized droplets with defined volume, geometry and position and a speed of up to 375 kHz. In this work, without using a superhydrophobic or superoleophobic surface, arrays of droplets are instantly formed on the patterned substrate by sliding a strip of liquid, including water, low-surface-tension organic solvents and solution, along the substrate. To precisely control the volume of the droplets, we systemically investigate the effects of the size of the wettable pattern, the viscosity of the liquid and sliding speed, which were found to vary independently to tune the height and volume of the droplets. Through this method, we successfully fabricated an oriented single metal-organic framework crystal array with control over their XY positioning on the surface, as characterized by microscopy and X-ray diffraction (XRD) techniques.

1. Introduction The generation of droplet arrays possesses great significance, both for the fundamental study of wetting[1, 2], dewetting[3, 4] and thermodynamics[5, 6] as well as various applications to compartmentalize chemical reactions[7, 8], biological analyses and materials syntheses[9, 10]. Many methods have been developed to fabricate droplets, such as laser microfabrication[11-14], inkjet printing[10, 15], microfluidic pen lithography[16, 17], electrowetting-ondielectric (EWOD) actuation[18,19], trapping droplets by microcavities[3], solvent exchange[20-22], and discontinuous dewetting on a patterned surface. Among these methods, discontinuous dewetting is a straightforward method to produce high-density microdroplet arrays at low cost by dewetting the liquid along a hybrid surface with regions of different wettability[2, 16]. Huizeng Li and co-workers fabricated femtoliter droplet arrays by splitting an aqueous mother drop on a patterned superhydrophilicsuperhydrophobic substrate, thereby isolating single cells[23]. Shaozhou Li et al. and Kerui Xu et al. deposited aqueous droplet arrays on a (super) hydrophilic/hydrophobic surface using the dip-coating method, respectively[24, 25]. Jin-Liang Zhuang et al. deposited droplet arrays of dimethylsulfoxide (DMSO) solution on Au substrates with patterned selfassembled monolayers by spin-coating[26]. Wenqian Feng et al. produced low-surface-tension liquid droplets arrays by moving a bulk drop on highly wettable and nonwettable slippery areas on smooth glass or flexible polymer films[27]. Using this method, which did not use a superoleophobic surface, a wide range of low-surface-tension liquids, including ethanol, methanol and even hexane, could be used to generate

Coorespondind author. E-mail Address: [email protected]

separated droplets arrays. These single-step formations of thousands of droplets in precise locations in an array format with desired volumes provide a unique solution for highthroughput applications. However, large-scale droplet arrays require uniform, precise and controllable dewetting speed and additional reagent consumption because the hybrid surface must first be wetted with bulk liquid. In this work, we developed a facile, ultra-high-throughput droplet array generation method by sliding a liquid strip along the substrate in a controllable manner with low reagent consumption. Millions of femtoliter to nanoliter droplets with defined positions, geometry and volume could be produced in seconds (up to 375kHz). This method enables the fabrication of droplet arrays compatible with polymer, oil, water, mixture and organic liquids with surface tension as low as γ lv=18.8 dyn/cm. In addition, the key factors in controlling the droplet volume, including the wettable pattern size, the viscosity of the liquid, and the relative sliding speed of the liquid strip, were also discussed systematically. These factors can also be regulated to control the volume of the droplet, which is significant for materials synthesis, especially for the oriented growth of single crystals. In addition to providing insights into the fundamental mechanism underlying the control of the droplet volume, we also demonstrate, as a proof-of-concept application, a simple technique to fabricate arrays of singlecrystal metal-organic frameworks (MOFs). This method is based on evaporation-induced self-assembly crystallization in the femtoliter droplet arrays.

2. Material and methods 2.1. Chemicals and materials AZ9260 photoresist, AZ400K developer and SU8 2002 photoresist were purchased from Suzhou Wenhao Microfluidic Technology Co., Ltd. (Suzhou, China). 1H,1H,2H,2H-Perfluorooctyltriethoxysilane (POTS) was

Please do not adjust margins

obtained from Sigma-Aldrich (USA), and silicone oil (η=10 mPa∙s, 25°C) was obtained from Hangping Company (Beijing, China). Cu(NO3)2, 1,3,5-benzenetricarboxylic acid (BTC), DMSO, glycerol (≥99.5%), coumarin 6, fluorescein sodium salt and hexane were purchased from Aladdin (Shanghai, China). Ethylene glycerol (≥99%) and linoleic acid were purchased from Sigma-Aldrich and Alfa Aesar, respectively. The other chemicals were all obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Deionized water (18.2 MΩ cm, S30CF, Master Touch, Shanghai, China) was used to prepare aqueous solutions. Silicon wafer substrates were purchased from Ningbo Sibranch International Trading Co., Ltd. (Ningbo, China). FC 40was purchased from 3M (China). 2.2. Fabrication of nonwettable substrates The glass substrates or silicon wafers were treated by being dipped into a mixed solution of saturated sodium hydroxide and ethanol for at least 24 hours. The substrates were washed with deionized water and then dried in an oven at 120°C for at least 10 hours. The substrates were further cleaned using an air plasma cleaner (PDC-002, Harrick Plasma), and then POTS was evaporated onto the substrates at 120°C for 1 hours in the oven (FD115, Binder, Germany).

is about 87°. After plasma treatment in 15 s, 30 s, 60 s with 30W RF power, all the receding angles reduced to 0°. Therefore, 2 min plasma treatment is long enough for the dewetting process (Fig.S2).The targeted wettable/nonwettable pattern was successfully obtained after removing the photoresist using acetone and ethanol sequentially. 2.4. Droplet arrays generation by the sliding method The droplet arrays are generated by a device (Fig.S1) that can be used to simultaneously fabricate millions of droplets. We used a piece of glass as the sliding tool and added 10 to 20 microliters of liquid into the slit between the sliding tool and the substrate to form a long liquid strip by capillary force. After adjusting the sliding speed and slit height, the liquid strip was pulled by a custom-designed stepping motor at constant speed. The liquid strip selectively adhered to the wettable spots surrounded by nonwettable areas to form arrays by discontinuous dewetting. The most prominent advantage of this method is the ability to control the sliding speed via mechanical manipulation, which is necessary for the controllable preparation and application of droplets. In addition, the template can then be readily reused (over 5 times in our experiments) for fabricating a new batch of droplets, enabling a rapid, inexpensive and easily reproducible method for fabricating droplet with various liquids.

2.3. Preparation of wettable patterns on the nonwettable surface 2.5. Characterization The nonwettable substrate was spin-coated with photoresist AZ9260 at 500 rpm for 6 seconds and subsequently at 2400 rpm for 60 seconds, followed by baking the coated substrate on a hotplate at the constant temperature of 105°C for 330 seconds. After baking, the nonwettable substrate was covered by a designed photomask and exposed to UV light (MJB4, SUSS MicroTech).Next, the patterned substrate was developed in an AZ400K aqueous solution (AZ400K: H2O=1:2) for 150 seconds and then washed with deionized water to remove the residual developer. Finally, the developed substrate was treated with plasma for two minutes, making the exposed area selectively wettable. We chose air plasma which not only remove the contamination, but oxidize the exposed surface. The receding angle is related to the plasma exposure time. We measured the receding angle under ambient conditions. For the nonwettable substrate, the receding angle

Fluorescent images were obtained using a fluorescence microscope (IX73, Olympus). Sectional images of the droplets were obtained using confocal laser scanning microscopy (FV3000, Olympus), and the heights of the droplets were measured using computer software (FV31S-SW, Olympus) that analyzed image slices taken along the Z-axis at intervals of 0.3-1.0 μm. The structures and morphology of the MOF crystal were investigated by using an X-ray diffraction (XRD, Panalytical Empyrean, Netherlands) apparatus, an optical microscope (LV100ND, Nikon), and a scanning electron microscope (SEM, GeminiSEM300,Zeiss, Japan). The viscosity of the liquids was measured using a circular-platetype viscometer (Haake mars 3, Thermo Scientific). The contact angle was measured by using a contact angle measuring instrument (XG-CAMC3, XYCXIE, Shanghai).

Fig.1. (a) Schematic illustration of the photolithography process of wettability patterning. (b) Schematic illustration of the sliding process, which splits droplets by sliding the liquid strip on the patterned substrate to form the droplet array. (c) Cross-sectional view of the sliding process. (d) The droplet arrays of DMSO on a chip. Scale bar is 2 cm.

3. Results and Discussion 3.1. Generation of droplet arrays by the sliding method A facile sliding method was developed to produce droplet arrays on a patterned substrate. The process included fabricating the patterned substrate (Fig.1a) and applying liquid by sliding it along the patterned substrate to generate droplet arrays. To prepare highly wettable patterns on the nonwettable surface, POTS molecules were selectively removed by air plasma after the lithographic process (Fig. 1a). Next, we used the droplet-generation device to fabricate open droplet arrays using the sliding method (Fig.1b, c). The sliding method has broad applications for various liquids with different surface tensions (from 18.8 to 72 dyn/cm), including polar liquids (ethylene glycol, glycerol), organic oil (silicone oil, oleic acid, linoleic acid), and photoresist (SU8 series) (Fig.2, Table S1). Fig.2 shows the droplets fabricated with different liquids via the sliding method. Note that the sliding method is useful for generating droplets with a defined shape and volume. In addition to simple geometries, such as hemispherical, we could also fabricate droplets with complex letters and number shapes (Fig.2a, b). Two special cases were obtained using our sliding methods. First, for water, a larger volume of water (at least 40 μL) must fill the slit to cover the entire patterned area to form droplets due to the hydrophobicity of the substrate. Second, for liquids with extremely low surface tension, such as FC40, our patterned substrate became fully wettable with no wettability contrast, resulting in weakened confinement and unsuccessful droplet formation. (a)

(b)

(c)

(d)

(e)

(f)

Fig.2. Various types of liquids used in the sliding method. (a) Glycerin. (b)Glycerin with fluorescence sodium. (c) Oleic acid. (d) Silicone oil. (e) Photoresist with coumarin 6, and (f) mixed solution of PdCl2 and linoleic acid. Scale bars are (a)(c)(f)500μm, (b) 300 μm, (d)(e) 200 μm.

Essentially, the sliding process is based on discontinuous dewetting. Dewetting, which is the reverse process of wetting, is a spontaneous process, and in this case, it causes a thin film or strip on a surface to rearrange into an ensemble of separated objects. Controlling the dewetting process enables obtaining objects with pre-defined statistical sizes and spatial distributions, starting from a homogeneous continuous system. To achieve the effect of discontinuous dewetting, specific regions must exhibit significant difference in wettability

(regions with good wettability make it easier for droplets to adhere, while those with good dewettability promote dewetting as the liquid moves)[3]. (a)

sliding

(c)

(b)

sliding

(d) sliding

wettable spot

sliding

nonwettable area

Fig.3. Illustration of the sliding process on pre-patterned substrates. (a) Bulk liquid dewetted from the surface at a constant receding angle. (b) The receding angle decreased at the boundaries between the wettable and nonwettable area. (c) The liquid was pinned to the boundary as the bulk liquid receded. The liquid drained away, thinning the strip,(d) The droplet was formed.

Due to the obvious difference in wettability between the wettable spots and the nonwettable barriers, the liquid is spontaneously removed from the barriers, and fills the wettable regions, thus generating high-density arrays of completely separated droplets. A liquid deposited on a substrate in air is governed by Young's equation:

 SL   AL cos    SA

(1)

Where γSL, γAL and γ SA are the substrate-liquid, air-liquid and substrate-air interfacial tensions, respectively, and θ is the contact angle between the liquid on the substrate in air constrained by γ SL and γ AL. In our case, when the liquid strip was pulled by the controllable mechanic force, the liquid retracted from the nonwettable surface at a constant receding contact angle (Fig. 3a). When the rear three-phase-contact line of the bulk liquid strip met the boundary of the wettable and nonwettable area, the rear contact angle decreased abruptly, thus pinning the three-phase-contact line. As the bulk liquid continued to be pulled, (Fig. 3b), the liquid film became increasingly thin (Fig. 3c)until a critical thickness was reached, at which point the liquid film ruptured and left a portion of the liquid droplet pinned on the wettable spot (Fig. 3d). 3.2. Volume control of the droplet To precisely control the volume of droplets, we systemically investigated the effects of the size of the wettable patterns (Fig. 4a), the viscosity of the bulk liquid, and the relative sliding speed between the liquid strip and the substrate (Fig. 5). Under the same conditions (sliding speed of 20mm/min, and the height of the liquid strip of 0.2mm), ten microliters of glycerol with

fluorescein sodium salt (0.09mM/L) was loaded into the slit, and then, glycerol droplet arrays were fabricated after sliding.

Fig. 4. (a) Effect of the size of the wettable spots on the volume of droplets prepared by sliding a glycerol strip (with fluorescein sodium salt) on a patterned substrate. (b-c) Top, sectional and 3D views of the droplets. The scale bars are 50 μm in (b) and 100 μm (c).

Fluorescent and 3D images of the obtained glycerol droplet arrays with different sizes are shown in Fig. 4. As the liquid strip moved, the droplet array was generated with wettable spots varying from 10 to 700 μm in diameter. In our experiment, we used three substrates on which the droplets are 10-105 μm, 40-400 μm and 200-700 μm in diameters. The measurement of the height was using three different substrates in all process. It is notable that the fabrication of droplets on different substrates is under the same experimental conditions. The heights of the droplets were measured using a confocal laser scanning microscope, and the droplet volumes were calculated using the radius(r) of the droplet at the base and the droplet height (h), as calculated by the spherical cap formula (2):

V

h 6

(3r 2  h 2 )

the solutions varied from 20 to 100 wt%, the volume of droplets correspondingly changed from 0.126 to 31.2 nL. Similarly, for fixed sliding speeds of 30, 20, 10, and 1 mm/min, as the mass fraction of water-glycerol solutions increased from 20 to 100 wt%, the droplet volumes increase from 0.126 to 27.33 nL, from 0.137 to 21.33 nL, from 0.126 to 18 nL, and from 0.142 to15.6 nL, respectively. The relationships between the volume of the droplet and the sliding speed and, viscosity are illustrated in Fig. 5. The volume of the droplets increased with the increase in sliding speed and viscosity because a higher sliding speed or viscosity increased the shear stress, corresponding to the sliding mother drop method on a patterned superhydrophobic surface[23]. To demonstrate the factors on which the volume of the generated droplets depended, we considered the effects of these key factors and found that the volume of the droplet increased with the increasing size of the wettable spots, viscosity of the bulk liquid, and relative sliding speed between the liquid strip and the substrate. Moreover, the relative sliding speed could be varied independently to tune the height and volume of droplets with the same liquid and spot size.

(2)

The height and the volume of the droplets ranged from 4.35 to 86.06 μm and, 213.81 fL to 16.885 nL (spanning six orders of magnitude), respectively. We regard the droplets as spherical caps because the profile of the droplets could be spherically fitted. We took a droplet of 700 μm in diameter as an example. Other size of droplet can also be fitted (Fig.S3). As shown in Fig.4(a), the volume of the glycerol droplets exponentially increased with the increasing lateral length of the wettable spots. Because such tiny amount of liquid could evaporate so quickly that will influence the measurements of height. We selected the droplets with 10 μm in diameter as an example to investigate the height stability in time. We measured three droplets height successively in four different positions and the measurement interval is 3 minutes. From the results, we find the maximum variation of droplet height and volume do not exceed 6% and 9% in 12 minutes. The time is long enough for us to measure the height using a confocal microscope (Fig. S4). To investigate the effect of the viscosity and sliding speed on the volume of droplets, four water-glycerol solutions with different glycerol contents were prepared and applied to produce droplets via the sliding method. The physical properties of the four water-glycerol solutions are summarized in Table 1[28, 29], demonstrating the obvious differences in viscosity between the four solutions. Ten microliters of liquid was slid onto the 200 μm patterned substrates. The sliding speed was set to 1, 10, 20, 30, and 40 mm/min, and the glycerol contents varied from 20 to 100 wt%. When the sliding speed was fixed at 40 mm/min, and the mass fraction of glycerol in

Fig.5. Effects of the viscosity and sliding speed on the volume of droplets prepared by sliding glycerol-water solutions with different mass fractions on 200 μm patterned substrates.

3.3. The growth of oriented single-crystal MOF arrays by sliding method The generation of droplet arrays on a patterned substrate via the sliding method is essential for a range of applications, especially for the controllable growth of crystals. Here, we show one such application where femtoliter droplets are used to generate MOF single crystals and to control the morphology and growth orientation accurately. MOFs are an class of porous crystalline materials with exceptionally high accessible surface areas due to their framework structure[30], which exhibit remarkable capabilities in gas storage, molecular separation,

sensing, controlled release and catalysis. We applied the sliding method to form HKUST-1 [Cu3(BTC)2] crystals, a well-studied MOF material consisting of metallic Cu(II) and BTC[31].We designed a mask with an array of more than 9 million circles, each 10 µm in diameter. Following this strategy, 9 million femtoliter-sized droplets of the HKUST-1 precursor solution could be fabricated in 24 seconds with an ultra-high throughput of up to 375 kHz using our method. Subsequently, monodisperse MOFs crystals grew upon evaporating the patterned precursor droplets to achieve oriented singlecrystallites via in situ crystallization. The obtained HKUST-1 crystals are shown in Fig. 6a and S5, demonstrating a single-crystal-per-micropatch resolution with hexagonal single crystals inside the micropatches via evaporation-induced self-assembly (EISA) on the patterned substrate. Interestingly, well-shaped, oriented single crystals were observed in the droplets with the desired size and position due to our controlled sliding method. The structures of the HKUST-1 crystals were confirmed via XRD analysis, as shown in Fig. 6b, showing the hexagonal single crystals preferentially orient along the [111] direction. The results agree well with those of other existing methods, and demonstrate the advantages of the developed method, including lower reagent consumption, higher throughput and facile fabrication. Moreover, the ability to fabricate droplets with controlled volumes, positions and geometries using our sliding method raises possibilities for future work on functional material synthesis. (a)

(b)

Fig.6. Characterization of HKUST-1 prepared from precursor solution via evaporation. (a) SEM images of HKUST-1 single crystals in 10 μm patches. Scale bars are 10 μm, and that of the inset of (a) is 2 μm, (b) XRD pattern of the on-chip crystallized HKUST-1 single crystals (black curve) with a standard pattern for HKUST-1for comparison (red curve).

4. Conclusions A facile, controllable and high-throughput method was successfully developed to fabricate droplet arrays with controlled size, geometry and position on a patterned surface. When we slid the liquid strip, a small volume of the liquid was confined to specific areas with higher surface energy than that of the background surface. Various liquids with different surface tensions (low to 18.8 dyn/cm) are applicable without requiring a superhydrophobic or superoleophobic surface. Arrays of separated droplets can be produced with low reagent consumption by using the sliding method on a patterned substrate. Droplets with volume ranging from femtoliters to nanoliters were also successfully achieved on one chip within seconds, which can facilitate meeting multiple requirements.

The volume of the droplets was well controlled by adjusting the size of the wettable pattern on the substrate, the viscosity of the liquid, and the sliding speed on the substrate. This method can be easily scaled-up to achieve an even higher throughput by sliding a longer liquid strip without interference. Using this sliding method, we successfully fabricated wellshaped, oriented single-crystal HKUST-1 arrays. We anticipate that this method will yield breakthroughs in the fabrication of patterned nanomaterials and self-assembled materials, and significantly promote promising applications in combinatorial chemistry, biological analysis, and related research. Currently, we are working towards more applications of this novel technology.

Acknowledgements The authors are grateful for financial support from the Shanghai Science and Technology Committee (Grant 16DZ2260601) and the Shanghai Pujiang Program (Grant 16PJ1403200)

References [1] A. Promraksa, Y.C. Chuang; L.J. Chen, Study on the wetting transition of a liquid droplet sitting on a square array cosine wave-like patterned surface, J Colloid Interf Sci. 418 (2014) 819. [2] A.A. Popova, K. Demir, T.G. Hartanto, et al., Dropletmicroarray on superhydrophobic-superhydrophilic patterns for high-throughput live cell screenings, RSC Adv. 6(2016)3826338276. [3] R.J. Jackman, D.C. Duffy; E. Ostuni, et al., Fabricating large arrays of microwells with arbitrary dimensions and filling them using discontinuous dewetting, Anal Chem. 70 (1998) 22802287. [4] Y. Wang, T.J. McCarthy, Dip-Coating Deposition on Chemically Patterned Surfaces a Mechanistic Analysis and Comparison with Topographically Patterned Surfaces, Langmuir. 30(2014)2419-2428. [5] D.I. Yu, H.J. Kwak., S.W. Doh, et al., Dynamics of Contact Line Depinning during Droplet Evaporation Based on Thermodynamics, Langmuir. 31 (2015)1950-1957. [6] D.S. Corti, K.J. Karl, K. Torabi, On the interfacial thermodynamics of nanoscale droplets and bubbles, J Chem Phys. 135 (2011). [7] H. Song, D.L. Chen, R.F. Ismagilov, Reactions in Droplets in Microfluidic Channels, Angew Chem Int Edit. 45(2006) 73367356. [8] J. Wu, M. Zhang, X. Li, et al., Multiple and High-Throughput Droplet Reactions via Combination of Microsampling Technique and Microfluidic Chip, Anal Chem. 84 (2012) 9689-9693. [9] Z. Chen, L. Hu, M. J. Serpe, Liquid-Liquid Interface Assisted Synthesis of Multifunctional and Multicomponent Hydrogel Particles. J. Mater. Chem., 22 (2012) 20998-21002. [10] M. Abulikemu, E. H. Da.as, H. Haverinen, et al., In Situ Synthesis of Self-Assembled Gold Nanoparticles on Glass or Silicon Substrates through Reactive Inkjet Printing. Angew. Chem., Int. Ed., 53 (2014)420-423. [11] B. Farshchian, J.R. Gatabi, S.M. Bernick, et al., Laser-induced superhydrophobic grid patterns on PDMS for droplet arrays formation, Appl. Surf. Sci. 396 (2017) 359-365. [12] M. Barberoglou, V. Zorba, E. Stratakis, et al., Bio-inspired water repellent surfaces produced by ultrafast laser structuring of silicon, Appl. Surf. Sci. 255 (2008) 5425-5429.

[13] S. Pelt, A. Frijns, R. Mandamparambil, et al., Local wettability tuning with laser ablation redeposits on PDMS, Appl. Surf. Sci. 303 (2014) 456-464. [14] P. Serra, J.M. Fernandez-Pradas, M. Colina, et al., Laserinduced forward Transfer: a Direct-writing Technique for Biosensors Preparation, J Laser Micro Nanoen. 1 (2006) 236242. [15] L. Zhang, J. Wu., M.N. Hedhili, et al., Inkjet printing for direct micropatterning of a superhydrophobic surface: toward biomimetic fog harvesting surfaces, J Mater Chem A. 3(2015) 2844-2852. [16] K.J. Bachus, L. Mats, H.W. Choi, et al., Fabrication of Patterned Superhydrophobic/Hydrophilic Substrates by Laser Micromachining for Small Volume Deposition and DropletBased Fluorescence, ACS Appl Mater Inter. 9 (2017) 76297636. [17] C. Carbonell, K.C. Stylianou, J. Hernando, et al., Femtolitre chemistry assisted by microfluidic pen lithography, Nat Commun. 4 (2013). [18] S.K. Cho, H. Moon; C.J. Kim, Creating, transporting, cutting, and merging liquid droplets by electrowetting-based actuation for digital microfluidic circuits, J Microelectromech S. 12 (2003)70-80. [19] H.W. Lu, F. Bottausci, J.D. Fowler, et al., A study of EWODdriven droplets by PIV investigation, Lab Chip. 8 (2008) 456461. [20] L. Bao, A.R. Rezk, L.Y. Yeo, et al., Highly Ordered Arrays of Femtoliter Surface Droplets, Small. 11 (2015)4850-4855. [21] S. Peng, D. Lohse, X. Zhang, Spontaneous Pattern Formation of Surface Nanodroplets from Competitive Growth. ACS Nano. 9(2015) 11916-11923. [22] B. Dyett, H. Yu, X. Zhang, Formation of surface nanodroplets of viscous liquids by solvent exchange, Eur Phys J E. 40 (2017). [23] H. Li, Q. Yang, G. Li, et al., Splitting a Droplet for Femtoliter Liquid Patterns and Single Cell Isolation, ACS Appl Mater Inter. 7 (2015)9060-9065. [24] S. Li, G. Lu, X. Huang, et al., Facile growth of a single-crystal pattern: a case study of HKUST-1, Chem Commun. 48 (2012) 11901-11903 [25] K. Xu, X. Wang, R.M. Ford, et al., Self-Partitioned Droplet Array on Laser-Patterned Superhydrophilic Glass Surface for Wall-less Cell Arrays, Anal Chem. 88 (2016) 2652-2658 [26] J.L. Zhuang, D. Ceglarek, S. Pethuraj, et al., Rapid RoomTemperature Synthesis of Metal-Organic Framework HKUST-1 Crystals in Bulk and as Oriented and Patterned Thin Films, Adv Funct Mater. 21 (2011) 1442-1447. [27] W. Feng, L. Li, X. Du, et al., Single-Step Fabrication of HighDensity Microdroplet Arrays of Low-Surface-Tension Liquids, Adv Mater. 28 (2016) 3202-3208 [28] J.B. Segur, H.E. Oberstar, Viscosity of Glycerol and Its Aqueous Solutions. Ind. Eng. Chem., 43 (1951) 2117–2120. [29] K. Takamura, H. Fischer, N.R. Morrow, Physical properties of aqueous glycerol solutions,J PETROL SCI ENG. 98-99 (2012)50-60. [30] S.L. James, Metal-organic frameworks, Chem Soc Rev. 32 (2003) 276-288. [31] C. Carbonell, I. Imaz, D. Maspoch, Single-Crystal MetalOrganic Framework Arrays, J Am Chem Soc. 133 (2011) 21442147.

Table 1. Viscosity, Surface Tension, Density of Aqueous Glycerol Solutions (20 °C) [28,29]

Glycerol (wt%)

Viscosity(mPa∙s)

Surface tension(dynes/cm)

Density (grams/ml)

100%

1412

63.4

1.00392

80%

60.1

67

1.01752

40%

3.72

70

1.03945

20%

1.76

71.7

1.0469