Flexible electronics based on magnetic printing and the volume additive principle

Dongdong Hu , Kaijing Zheng , Feng Yang , Jun Nie and Xiaoqun Zhu *
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, P. R. China. E-mail: zhuxq@mail.buct.edu.cn; Fax: +86 10-6442-1310; Tel: +86 10-6442-1310

Received 5th June 2017 , Accepted 17th July 2017

First published on 27th July 2017

Microscale and nanoscale magnetic objects can be driven and assembled in defined locations efficiently and rapidly in magnetic fields. In this study, we present a strategy to fabricate flexible electronic circuits using non-contact magnetic printing and a volume additive substitution reaction. The process involves assembling iron nanoparticles in designated fields in the magnetic field, thereby forming patterns of iron nanoparticles. Thereafter, the nanoparticles were fused together by converting the iron nanoparticles to silver through an Fe–Ag replacement reaction. A silver flexible electronic circuit was fabricated conveniently through this process. Using this approach, flexible electronic circuits with double-sided conductive patterns could be prepared. This method provides a general and highly effective approach to fabricate various conductive silver patterns on flexible materials, such as paper, PI, PET, and others.

1. Introduction

The invention of flexible electronics has caused tremendous developments in modern electronics. Owing to the flexibility, thin layer, low mass, and bending resistance1 of flexible electronics, they are widely used in flexible transistors,2 antennas,3 solar cells,4 radio frequency identification,5 flexible displays,6 wearable electronics,7,8 and sensors.9 As expected, due to the commercial prospects of flexible electronics, fabricating these circuits is becoming popular in academic research and industrial applications. Therefore, developing facile and eco-friendly preparation methods for flexible electronics has become necessary.

Commonly, flexible electronics are manufactured through electroplating and etching, which are time consuming, complicated, and have highly hazardous pollution effluents. Hence, numerous attempts have been focused on developing new techniques for manufacturing flexible electronics. Among these approaches, printed electronic technologies, which include screening printing,10 gravure,11 flexography,12 and inkjet printing,13 have been the most advantageous techniques in recent years. The main conductive materials of the printing inks can comprise intrinsically conductive polymers such as PEDOT:PSS14 or polyaniline,15 or carbon-based materials such as nanotubes or graphene and graphene composites16 that do not require any other further heat treatment but feature a much smaller conductivity and therefore cannot be applied to power transmission lines but rather to sensing lines. Thus, inks with metallic nanoparticles that could provide good conductivity have been widely researched and have been applied in some fields. Such inks are composed of metallic nanoparticles and carrier liquid solvents. However, the size of the metal powder in the ink is difficult to control and stabilize during transportation and application.17 Moreover, the coffee ring effect should also be considered in a solvent-based ink, which should be optimally formulated18 or/and a different printing strategy should be used, such as switching to the Print On Fly (POF) mode,19 to avoid patterns with a coffee ring effect. Owing to electrostatic interaction, the problem of misting is inevitable. A large amount of dispersant and other organic compounds must be added to address the above problems and to increase the conductivity, the nanoparticle-based inks require sinterization as a post-printing treatment.20 Therefore, numerous interesting and low-cost substrate materials, such as thermoplastic polymers or paper, cannot be used.21 Thus, for controlling metal powder size, antioxidation, and stable dispersion of the metal powder, coffee rings involved in electronic printing technologies drastically hamper large-scale production and applications. For future flexible electronic devices, developing a novel surface metallization strategy that directly deposits metallic circuits on flexible substrates would be useful and cost effective.

Previously, we reported a two-step method to prepare flexible electronics via photoreduction22–25 and a volume additive approach in situ.22 Using the “volume additive process on demand” theory based on the replacement reaction, which uses a substitutional metal with a large atomic radius to replace the seed metal with an atom with low radius to eliminate the interspace among the metal nanoparticles of the pattern, the pattern formed by the metal particles could be fused together and become easily conductive. The photoreduction of the metal ion offers the advantage of spatial control, but the reaction of photoreduction remains slow for industrial applications; moreover, the metal film thickness is confined to the limited light penetration into the color solution. Thus, a new strategy for the rapid formation of metal patterns and a tunable pattern thickness is required to solve the above problems.

In 1973, J. Watson proposed the theory of trapping magnetic particles. Since then, magnetically assembling particles has been regarded as a powerful tool for rapidly organizing nanoscale matter into complex structures.26,27 Magnetic field microgradients established using “virtual modulators” that act as templates to assemble both non-magnetic and magnetic particles have been used for numerous studies.28,29 These templates with special patterns are made of high-magnetic permeability materials (HMPMs, such as iron) and low-magnetic permeability materials (LMPMs, such as aluminum, air, and polymer).30,31 When the modulator is placed in a uniform magnetic field, the magnetic indication flux tends to go through HMPMs and bypass LMPMs, thereby leading to a nonuniform magnetic field. Moreover, magnetic flux density increases above HMPMs and decreases above LMPMs when the directions of the magnetic indication flux and magnetic field are vertical. Thus, if the paramagnetic particles are jetted onto the modulator surface in the magnetic field, the particles would be attracted onto the HMPMS material regions.

Therefore, in this research, we propose a novel strategy based on magnetic-assisted assembly and a volume additive process to manufacture flexible electronic circuits. Iron nanoparticles were magnetically assembled into a given pattern. However, this pattern was accumulated by loose particles, resulting in a non-conductive pattern. Here, the volume additive process could help facilitate conductivity. Through the in situ replacement reaction of silver–iron, the iron pattern was converted into a conductive silver pattern. One iron atom could reduce three silver ions; thus, the metal volume would increase substantially. The previously loose particles were fused together through the Fe–Ag replacement reaction. Finally, a conductive silver pattern was achieved. The whole process is shown in Scheme 1.

image file: c7tc02476b-s1.tif
Scheme 1 Fabrication of the conductive pattern. (a) Coating the UV adhesive on the substrate; (b) the adhesive is cured by illumination using a 365 nm LED lamp with the intensity of 30 mW cm−2; (c) the ethanol suspension with iron nanoparticles is sprayed and trapped by the magnetic field; (d) the oxygen inhibition layer is cured completely by illumination with the intensity of 30 mW cm−2 to fix the pattern on the substrate; (e) the substrate with iron nanoparticles is immersed in an ethanol solution of AgNO3, and iron is replaced by silver through the replacement reaction; (f) the silver pattern.

2. Experimental

2.1 Materials

All the reagents, namely, ferric chloride hexahydrate, sodium borohydride, and silver nitrate, were of analytical grade. Ferric chloride hexahydrate was purchased from Alfa Aesar. Sodium borohydride was purchased from Aladdin. Silver nitrate was purchased from Sinopharm Chemical Reagent Beijing Co., Ltd.

2.2 Magnetic printing of the iron nanoparticle pattern

We prepared and preserved the iron nanoparticles to reduce oxidation. The preparation and size distribution of the iron nanoparticles are shown in ESI S1. The fabrication of the magnetic modulator is demonstrated in ESI S2.

Prior to being placed in the magnetic field, the flexible substrate was coated with a 30 μm UV coating, which was used to bind the metal pattern with the substrate (the coating formula and its curing operation are available in the ESI S3). Owing to the oxygen inhibition toward photopolymerization, the bottom of the coating was cured completely, and the coating was bound with the substrate well. Moreover, the surface of the coating was sticky enough to capture iron nanoparticles. After the particles adhered to the sticky surface, the surface was cured again under UV irradiation, and the particles adhered to the substrate.

The modulator was placed in a 20 mT uniform magnetic field. The flexible substrate was placed on the modulator. A droplet of alcohol suspension containing Fe nanoparticles was dropped onto the substrate. Fe nanoparticles self-assembled on the substrate quickly, in a pattern the same as that of the modulator. After the alcohol evaporated, the substrate was subjected to a second irradiation, and the Fe nanoparticles were bound to the substrate.

2.3 In situ volume addition based on the replacement reaction between Ag+ and Fe

The replacement reaction was done by immersing the substrate bearing the Fe nanoparticle pattern into an ethanol solution of silver nitrate for 10 min at pH = 2.73. Then, the sample was washed with deionized water and dried in a vacuum oven at room temperature.

2.4 Characterization

Iron nanoparticles were characterized using X-ray diffraction (XRD) with a wide-angle XRD analyzer (WAXD, D/max 2500 VB2t/PC, 40 kV, 100 mA) and monochromic Cu Kα radiation (λ = 0.154 nm). The morphology and pattern of the iron nanoparticles were investigated with a Zeiss Supra55 scanning electron microscope with an accelerating voltage of 20 kV. The silver pattern conductivity was characterized using an AVO meter (UT61E, Changsha Tai Shi Instrument Equipment Co., Ltd).

3. Results and discussion

3.1 Modulation of the magnetic-assisted assembly of iron nanoparticles in the magnetic field

In 1973, J. Watson proposed a ratio of two velocities, Vm/V0, to judge whether magnetic particles could be trapped in a magnetic field.29Vm/V0 > 1 demonstrates that magnetic particles could be trapped, and a high Vm/V0 ratio indicates a strong magnetic force acting on the magnetic particles. This ratio is given by
image file: c7tc02476b-t1.tif(1)
where Vm is the velocity of the magnetic particles caused by magnetic force when the nanoparticles are dispersed in the base load fluid, V0 is the velocity of random motion of nanoparticles in the base load fluid without a magnetic field, χ is the magnetic susceptibility of the magnetic particles, Ms is the saturation magnetization of the magnetic medium (here, iron nanoparticles), H0 is the intensity of the external magnetic field, η is the dynamic viscosity of the magnetic fluid, a is the width of magnetic medium wires, and b is the radius of the magnetic particles. In our research, Vm/V0 ≈ 128, which means that the magnetic force was 128 times higher than the other force. This magnetic force is substantially stronger than the other forces in our research (the calculation is shown in ESI S4). Thus, in the following simulation, only the magnetic force will be considered to simplify the simulation.

In this simulation, we considered the iron–aluminum fringe-patterned templates as examples to simulate magnetic field distribution. As shown in Fig. 1(a), the modulator was prepared using iron foils with 300 μm thickness and aluminum foils with 300 μm thickness. The modulator was 1 cm wide, 1 cm high, and 6 cm long. The external magnetic field was 20 mT, and the calculations for the magnetic flux density distribution (B) are shown in Fig. 1(b–d). As expected and as seen in Fig. 1(b), the magnetic flux density above the iron foils, which are HMPMs, was markedly higher than that of the above aluminum foils, which are LMPMs. As height increased, the magnetic field density over the two materials changed. During simulation, all the heights over the modulator remained consistent with the practical applications of substrate thicknesses of 100, 200, and 300 μm; however, a 30 μm thick UV adhesive was on the substrate.

image file: c7tc02476b-f1.tif
Fig. 1 Fringe pattern magnetic modulators made of iron and aluminum and the distribution of magnetic field around fringe patterns exposed to a uniform vertical external magnetic field of 20 mT. (a) Fringe pattern magnetic modulator made of iron and aluminum and placement direction in a uniform magnetic field in the vertical direction; (b) the change in simulated magnetic field density above the iron and aluminum fields with different heights; (c) simulated cross-section magnetic field distribution and plots of magnetic strengths at different heights over the fringe pattern; (d) simulated 3D cross-section magnetic field distribution over the fringe pattern magnetic modulator.

After calculating the distribution of the magnetic flux density over the fringe modulator, the force analysis of the as-prepared iron nanoparticles on the fringe modulator should be demonstrated. To simplify the analysis of the formation of the Fe nanoparticle pattern, the nanoparticles were dispersed in ethanol. The force on an iron nanoparticle inside a magnetic field depends on the strength and gradient of the applied magnetic field and could be expressed by32

image file: c7tc02476b-t2.tif(2)
where V [m3] is the volume of iron nanoparticles, χ1 is the magnetic susceptibility, χ2 is the magnetic susceptibility of the iron nanoparticles suspended in the ethanol, Δχ = χ2χ1, μ0 is the vacuum permeability, B is the magnetic flux density of the external magnetic, and ∇ is the Hamiltonian operator. Thus, to the iron nanoparticles in a definite magnetic field, the (V·Δχ)/μ0 is a constant. The force is only relative to (B·∇)B. The force on the iron nanoparticles with different magnetic flux densities of external magnetism and different thicknesses of the substrate was calculated, and the results are shown in Fig. 2. In reality, the height over the modulator includes the substrate thicknesses, namely, 100, 200, and 300 μm, and the UV adhesive thickness, which is 30 μm; therefore, the heights over the modulator in the calculation were 130, 230, and 330 μm. In Fig. 2, the magnetic force Fm values on the particles in the x- and z-axes are expressed as Fx and Fz respectively. Each iron nanoparticle suspended above the magnetic modulator was driven by both forces. The direction of Fx starts from the two sides to the center; therefore, the iron nanoparticles tend to move to the center of the HMPMs and LMPMs. Above the HMPMs (iron), the directions of Fz on the iron nanoparticles were downward and upward above the LMPMs (aluminum). Regardless of whether with HMPMS or LMPMS, the distribution of Fz over the modulator showed a wave shape, and the force decreased from the center to the two sides. Compared with that in Fz, the change in magnitude of Fx with different external magnetic flux densities and different heights was small. Thus, an increased external magnetic field indicated greater Fx and Fz, and the iron nanoparticles tend to be excluded to the region above LMPMS and were deposited to the middle of the HMPMs. Moreover, as shown in Fig. 2(d and e), for the increase in height over the modulator, the magnetic flux density decreased and the force declined; therefore, the substrate thickness should be considered during the self-assembly of iron nanoparticles.

image file: c7tc02476b-f2.tif
Fig. 2 Curves of calculated force on the as-prepared iron nanoparticles with different external magnetic flux densities and different substrate thicknesses.

3.2 Formation of an Fe nanoparticle pattern through magnetic self-assembly

As calculated, the iron nanoparticles could self-assemble into a fringe pattern on the flexible substrate, which was situated on the fringe modulator in a sufficiently strong uniform external magnetic field. Experimentally, to verify the modulation result, the effects of different external magnetic field intensities and different substrate thicknesses over the modulator on the pattern morphology were investigated. The formed patterns were observed using SEM and the results are shown in Fig. 3. Fig. 3(a–c) show the morphology of the Fe nanoparticle patterns with 5, 10, and 20 mT external magnetic field intensities, respectively; and the substrate thickness was 100 μm. Fig. 3(c–e) show the morphology of the Fe nanoparticle pattern on the substrate with thicknesses of 100, 200, and 300 μm at 20 mT external magnetic field intensity. In correspondence with the theoretical calculation (Fig. 2), at the external magnetic field intensity of 5 mT, (B·∇)B in the x- and y-axes were only 2.5 and 50 [mT (mT mm−1)], respectively, which were weak to drive the iron nanoparticles to self-assemble into the pattern. Thus, the pattern had a hazy outline and the fringes were connected to each other; moreover, the fringe was markedly wider than the modulator (the width of the iron in the modulator is 300 μm). As the external magnetic flux density increased, the (B·∇)B in the x- and y-axes were increased. At the external magnetic flux density of 20 mT, the values of (B·∇)B in the x- and y-axes were 40 and 550 [mT (mT mm−1)]. Under this condition, the pattern was clear and its width was equal to that of the modulator fringe. Ultimately, the influence of different substrate thicknesses on the pattern morphology was caused by the different magnetic field intensities. As calculated above (Fig. 2), a thick substrate results in a weak magnetic field intensity. Thus, as substrate thickness increased, the iron nanoparticle pattern clarity declined. These results agreed with those in the above simulation.
image file: c7tc02476b-f3.tif
Fig. 3 SEM images of Fe nanoparticle patterns under different magnetic flux densities at a height of 130 μm over the modulator, namely, (a) 5, (b) 10, and (c) 20 mT; and different heights over the modulator with the 20 mT external magnetic field at (c) 130, (d) 230, and (e) 330 μm.

3.3 Construction of a conductive silver pattern by a volume additive process in situ by replacement reaction of Fe–Ag+

As shown in Fig. 4a, the iron nanoparticle pattern was prepared successfully. However, the pattern was not conductive mainly owing to the discontinuous and loose accumulation of iron nanoparticles. Moreover, the surface of the iron nanoparticles was supposed to be oxidized, which would block the conductivity. This theory could be proven by the pattern morphology. SEM images were taken from the surface and cross section of the iron nanoparticle pattern (Fig. 4(b and c)). Regardless of the direction, the interspace was observed between iron nanoparticles. Even after compression (Fig. 4d), the interspace could not be eliminated. Therefore, to achieve the conductivity of the pattern, the key issue was to connect the metal particles with each other.
image file: c7tc02476b-f4.tif
Fig. 4 SEM images of the Fe nanoparticle and Ag patterns; (a) photo of the Fe nanoparticle pattern; (b) morphology of the Fe pattern, (c) cross-sectional morphology of the Fe nanoparticle pattern; (d) surface image of the Fe nanoparticle pattern after pressing; (e) photo of the Ag pattern; (f) morphology of the Ag pattern; (g) cross-sectional morphology of the Ag pattern; and (h) the surface image of the Ag pattern after being pressed.

Our previous work proposed a volume additive process to achieve a metal pattern with conductive nanoparticles.22 The volume additive process was based on the replacement reaction, which uses metals (substitutional metal) with large atomic radius to substitute the metal (seed metal) with small atom radius so as to eliminate the interspaces between the metal nanoparticles. In this research, this method was used to test the conductivity of the iron nanoparticle pattern because one atom of iron could reduce three silver ions, which could increase the metal volume and occupy the interspaces between the iron nanoparticles. After the substrate was immersed in the ethanol solution with silver nitrate, the Ag pattern was not enlarged and virtually had the same width as the iron pattern. This finding is attributed to the replacement reaction that only occurs in places with iron nanoparticles; therefore, the silver pattern could be kept in the pattern of the iron. In comparing the morphologies of the iron pattern with the silver pattern using the surface and cross section (Fig. 4(b, c, f and g)), the silver pattern microstructure was more compact. In particular, the silver pattern thickness was significantly increased relative to the iron pattern. These results demonstrated that the volume additive method worked in this research. This manner of volume addition showed an automatic fusion effect, which helped the metal pattern by having a continuous internal microstructure. This process could be demonstrated clearly by the schematic shown in Scheme 2.

image file: c7tc02476b-s2.tif
Scheme 2 Schematic of the Fe–Ag replacement reaction and the elimination of interspaces between the iron nanoparticles.

When a silver pattern on the flexible substrate is prepared using magnetic printing and a volume additive process, the conductivity should be verified. The conductivity of the silver pattern was tested and the result is shown in Fig. 5. The conductivity of the silver pattern was approximately 1.25 × 10−6 S m−1. Moreover, in a flexible substrate, the conductivities under different bending times, even at 100 times, remained virtually the same. These results proved that the flexible silver pattern fabricated using this two-step method offers good conductivity and good flexibility.

image file: c7tc02476b-f5.tif
Fig. 5 Conductivity of the Ag pattern and its bending conductivity.

As shown in Fig. 6, a series of flexible electronic circuits with different sizes and shapes were easily fabricated. Each circuit possessed good conductivity and the light-emitting diode lamps remained bright even when the pattern was bent. Moreover, using the magnetic printing and volume additive methods, a double-faced pattern on one substrate with different patterns was easily fabricated, as shown in (Fig. 6e and f). In Fig. 6f, the top side was illuminated using a white lamp and the bottom side was illuminated using a blue lamp. This finding presented that each side pattern exhibited good conductivity. These results demonstrated the simplicity and efficiency of our method to fabricate a large-scale flexible circuit and a double-sided board.

image file: c7tc02476b-f6.tif
Fig. 6 Conductive patterns and their usage.

4. Conclusion

In this paper, we presented a two-step method based on magnetic printing and the volume additive principle to prepare flexible electronics. The prepared flexible electronics possessed good conductivity and flexibility. This method is, therefore, suitable for numerous kinds of flexible substrates. Aside from silver electronics, copper electronics are expected to be fabricated using this method. This technology presents a potential application for electronic devices.


This study was supported by National Natural Science Foundation of China (No. 51603007, 51373015, and 51573011). The authors would also like to thank the Beijing Laboratory of Biomedical Materials for the support.


  1. T. George, M. S. Islam, A. K. Dutta, S. W. Bedell, D. Shahrjerdi, K. Fogel, P. Lauro, C. Bayram, B. Hekmatshoar, N. Li, J. Ott and D. Sadana, Proc. SPIE, 2014, 9083, 90831G Search PubMed .
  2. J. Zhao, W. Chen, J. Meng, H. Yu, M. Liao, J. Zhu, R. Yang, D. Shi and G. Zhang, Adv. Electron. Mater., 2016, 2, 1500379 CrossRef .
  3. T. Inui, H. Koga, M. Nogi, N. Komoda and K. Suganuma, Adv. Mater., 2015, 27, 1112–1116 CrossRef CAS PubMed .
  4. M. Ye, X. Hong, F. Zhang and X. Liu, J. Mater. Chem. A, 2016, 4, 6755–6771 CAS .
  5. Y. Zheng, Z. He, Y. Gao and J. Liu, Sci. Rep., 2013, 3, 1786 CrossRef .
  6. Q. Zhang, Y. Di, C. M. Huard, L. J. Guo, J. Wei and J. Guo, J. Mater. Chem. C, 2015, 3, 1528–1536 RSC .
  7. X. Liao, Z. Zhang, Q. Liao, Q. Liang, Y. Ou, M. Xu, M. Li, G. Zhang and Y. Zhang, Nanoscale, 2016, 8, 13025–13032 RSC .
  8. Y. Li, Y. A. Samad, T. Taha, G. Cai, S.-Y. Fu and K. Liao, ACS Sustainable Chem. Eng., 2016, 4, 4288–4295 CrossRef CAS .
  9. N. Münzenrieder, D. Karnaushenko, L. Petti, G. Cantarella, C. Vogt, L. Büthe, D. D. Karnaushenko, O. G. Schmidt, D. Makarov and G. Tröster, Adv. Electron. Mater., 2016, 2, 1600188 CrossRef .
  10. M. Wang, X. Xu, B. Ma, Y. Pei, C. Ai and L. Yuan, RSC Adv., 2014, 4, 47781–47787 RSC .
  11. R. Kitsomboonloha, S. J. Morris, X. Rong and V. Subramanian, Langmuir, 2012, 28, 16711–16723 CrossRef CAS PubMed .
  12. C. O. Phillips, S. Govindarajan, S. M. Hamblyn, R. S. Conlan, D. T. Gethin and T. C. Claypole, Langmuir, 2012, 28, 9878–9884 CrossRef CAS PubMed .
  13. N. N. Jason, W. Shen and W. Cheng, ACS Appl. Mater. Interfaces, 2015, 7, 16760–16766 CAS .
  14. A. Chiolerio, P. Rivolo, S. Porro, S. Stassi, S. Ricciardi, P. Mandracci, G. Canavese, K. Bejtka and C. F. Pirri, RSC Adv., 2014, 4, 51477–51485 RSC .
  15. A. Chiolerio, S. Bocchini, F. Scaravaggi, S. Porro, D. Perrone, D. Beretta, M. Caironi and C. F. Pirri, Semicond. Sci. Technol., 2015, 30, 104001 CrossRef .
  16. R. Giardi, S. Porro, A. Chiolerio, E. Celasco and M. Sangermano, J. Mater. Sci., 2013, 48, 1249–1255 CrossRef CAS .
  17. J. A. Lim, W. H. Lee, H. S. Lee, J. H. Lee, Y. D. Park and K. Cho, Adv. Funct. Mater., 2008, 18, 229–234 CrossRef CAS .
  18. M. Singh, H. M. Haverinen, P. Dhagat and G. E. Jabbour, Adv. Mater., 2010, 22, 673–685 CrossRef CAS PubMed .
  19. K. Rajan, S. Bocchini, A. Chiappone, I. Roppolo, D. Perrone, M. Castellino, K. Bejtka, M. Lorusso, C. Ricciardi, C. F. Pirri and A. Chiolerio, Flex. Print. Electron., 2017, 2, 024002 CrossRef .
  20. K. Rajan, I. Roppolo, A. Chiappone, S. Bocchini, D. Perrone and A. Chiolerio, Nanotechnol., Sci. Appl., 2016, 9, 1–13 Search PubMed .
  21. S. Ye, A. R. Rathmell, Z. Chen, I. E. Stewart and B. J. Wiley, Adv. Mater., 2014, 26, 6670–6687 CrossRef CAS PubMed .
  22. B. Wang, D. Hu, J. Nie and X. Zhu, Mater. Lett., 2017, 188, 296–299 CrossRef CAS .
  23. X. Zhu, B. Wang, F. Shi and J. Nie, Langmuir, 2012, 28, 14461–14469 CrossRef CAS PubMed .
  24. B. Wang, S. Chen, J. Nie and X. Zhu, RSC Adv., 2014, 4, 27381–27388 RSC .
  25. Y. Zhou, T. Ping, I. Maitlo, B. Wang, Y. A. Muhammad, J. Nie and X. Zhu, Nanotechnology, 2016, 27, 1–9 Search PubMed .
  26. R. M. Erb, H. S. Son, B. Samanta, V. M. Rotello and B. B. Yellen, Nature, 2009, 457, 999–1002 CrossRef CAS PubMed .
  27. D. Zerrouki, J. Baudry, D. Pine, P. Chaikin and J. Bibette, Nature, 2008, 455, 380–382 CrossRef CAS PubMed .
  28. A. F. Demirors, P. P. Pillai, B. Kowalczyk and B. A. Grzybowski, Nature, 2013, 503, 99–103 CrossRef PubMed .
  29. L. He, M. Wang, Q. Zhang, Y. Lu and Y. Yin, Nano Lett., 2013, 13, 264–271 CrossRef CAS PubMed .
  30. Y. S. Tsunehisa Kimura, F. Kimura, M. Iwasaka and S. Ueno, Langmuir, 2005, 21, 830–832 CrossRef PubMed .
  31. F. K. Guangzhe Piao and T. Kimura, Langmuir, 2006, 22, 4853–4855 CrossRef PubMed .
  32. N. Pamme, Lab Chip, 2006, 6, 24–38 RSC .


Electronic supplementary information (ESI) available. See DOI: 10.1039/c7tc02476b

This journal is © The Royal Society of Chemistry 2017