A magnetic nano-composite soft polymeric membrane

Microsyst Technol (2013) 19:409–418 DOI 10.1007/s00542-012-1646-2

TECHNICAL PAPER

A magnetic nano-composite soft polymeric membrane Akanksha Singh • Mandar Shirolkar • Mukta V. Limaye • Shubha Gokhale • Chantal Khan-Malek • Sulabha K. Kulkarni

Received: 30 September 2011 / Accepted: 31 July 2012 / Published online: 12 August 2012 Ó Springer-Verlag 2012

Abstract There is an increased need for low cost actuation technologies at the micro and nanoscale. Magnetically responsive polymer-based materials are good candidates for numerous applications in microsystems for actuation and sensing purposes. In this work, we report on nanopolymer composite magnetic silicone-based membranes, which provide the low elastic modulus needed for magnetic actuation to be effective at small scales. Passivated crystalline cobalt (*37 nm) and water based iron/cobalt (*100 nm) nanoparticles (NPs) have been synthesized using a chemical route at 50 °C and at room temperature, respectively. The NPs were characterized by Fourier Transform Infrared Spectroscopy, X-Ray Diffraction, Atomic Force Microscopy and Vibrating Sample Magnetometry (VSM). The NPs are then uniformly dispersed in a polydimethyl siloxane (PDMS) polymer matrix in order to fabricate smooth and flexible magnetic composite membranes. The magnetic properties of the membranes for different amounts of cobalt and iron NPs (16 and 25 wt%) A. Singh (&)  C. Khan-Malek Department Micro Nano Sciences and Systems, FEMTO-ST Institute, UMR CNRS 6174, MN2S, 32 Avenue de l’Observatoire, 25044 Besanc¸on Cedex, France e-mail: [email protected] URL: http://www.femto-st.fr/ M. Shirolkar  M. V. Limaye Department of Physics, University of Pune, Pune 411007, India S. Gokhale School of Sciences, Indira Gandhi National Open University, New Delhi 110068, India S. K. Kulkarni DST Unit on Nano Science, Indian Institute of Science Education and Research, Pashan, Pune 411021, India

were characterized by VSM and deflection measurements. Co/Fe PDMS composite membranes of about 50 mm diameter and *250 lm thickness were used under the application of *400 Oe magnetic fields. The cobaltPDMS membrane shows the largest deflection (*900 vs. *80 lm for an iron-PDMS membrane). The deflections observed on these membranes are found to have a linear dependence on the applied magnetic field.

1 Introduction Nanostructure forms of various metals, semiconductors, insulators, alloys and magnetic materials exhibit interesting properties, which make them potential candidates for building small size devices by forming nanocomposites with different polymers. Polymer-nanocomposites have been widely investigated which unify the advantages of polymers, like flexibility and inertness, while the nanomaterials contribute to the advantages due to their size dependent properties such as optical, magnetic, electrical, etc. (Wilson et al. 2004; Abyanch et al. 2007; Abyanch and Kulkarni 2008). In this work, we report on highly deformable magnetic membranes based on polymer-metal nanocomposites of cobalt and iron nanoparticles (NPs) dispersed in polydimethyle siloxane (PDMS). Magnetic particles dispersed in the polymer matrix have been already investigated (Khoo and Liu 2001; Yamahata et al. 2005; Yufeng et al. 2006; Berthier and Ricoul 2004; Park et al. 2004; Fahrni et al. 2009). However, there are few reports in which magnetic NPs have been used for fabricating magnetic membranes (Berthier and Ricoul 2004; Park et al. 2004). Ferromagnetic membranes showing large and reproducible deflection that can be operated under relatively low magnetic field are desirable for many applications.

123

410

Berthier and Ricoul have reported magnetic membranes using iron particles of *4 lm size and iron oxide (maghemite c-Fe2O3) particles of 10 nm size dispersed in PDMS (Berthier and Ricoul 2004). The deflections observed in their case were *200–300 lm with a nonlinear response to the magnetic field. Recently magnetic membranes using commercial NPs have been fabricated by Fahrni et al. (2009). They have used super paramagnetic iron-oxide NPs (10 nm) with a 5 % concentration by volume in PDMS. They failed to do the fabrication of membranes more than 5 % by volume because of the cluster formation in the PDMS matrix. They studied the mechanical properties of the fabricated membrane and showed that the Young’s modulus decreased with increasing particle concentration. They also described the magnetic behaviour of the composite polymer. Pirmoradi et al. fabricated composite membranes embedding coated iron oxide NPs in PDMS matrix (Pirmoradi et al. 2010). Commercial iron oxide NPs of Fe3O4/c-Fe2O3 (50/50 mixture) of 10 nm and uncoated iron oxide of 20–30 NPs diameter were used for the membrane fabrication with different thickness from 4 to 7 mm. They coated these commercial NPs with fatty acid to overcome the agglomeration problem of oxide particles in polymer. They achieved *625 lm deflection under *0.417 Tesla of magnetic field with 40 % w/w loaded NPs in PDMS. Particles with 25 % w/w loaded in PDMS had not significantly super paramagnetic property observed. Magnetic NPs have been widely studied, because of their excellent magnetic properties and potential applications such as microfluidics, ferrofluids, data storage devices, biomedicine like biomolecular separation, magnetic resonance imaging, targeted drug delivery, DNA separation and a.c. magnetic field-assisted cancer therapy, etc. (Pan et al. 2005; Yang et al. 2009; Pan et al. 2004; Rosensweig 1985; Speliotis 1999; Doyle et al. 2002; Gupta et al. 1988; Jordan et al. 1999; Oswald et al. 1995; Perez et al. 2003). Pan et al. fabricated magnetic actuated PDMS micropumps. Micropump consisted of valve and actuation chamber having permanent magnetic disc of neodymiumiron-boron (3.1 mm diameter and 1.6 mm thickness) in between and operated by external magnetic force of 500 Gauss. Yang et al. used ferrite magnetic particle (*25 lm) for their work. In this work they injected the particles in PDMS matrix with different concentration (10, 20, 30 %) under magnetic field of 1,500 Gauss. They studied magnetic field strength, structure formation under rotating magnetic field and actuation of these composite structures. PDMS peristaltic micropump with magnetic drive was fabricated by Pan et al. Three layers of PDMS were fabricated for micro pump in which they used permanent rods of Ni-plated-NdFeB (1.5 mm diameter) on the top. They studied deflection of membrane and fluid flow

123

Microsyst Technol (2013) 19:409–418

using numerical analysis. DC micromotor (6 mm diameter and 15 mm length) were used externally to actuate the internal magnetic rods of NdFeB. The metallic magnetic NPs have larger magnetization compared to metal oxide ones, which is interesting for many applications. But metallic magnetic NPs are not air stable, and are easily oxidized, resulting in the change or loss (fully or partially) of their magnetization. Therefore, strong capping is essential for these NPs. It is evident that there is a need for soft magnetic materials that can be used to obtain adequate magnetic properties (e.g., high permeability, high saturation magnetization, etc.). This satisfies a need for devices that are smaller in size to reduce costs and improve overall system efficiency. Cobalt NPs are known to exist in three polymorphs, the face-centered cubic (fcc), hexagonally close packed (hcp), and epsilon (e) phases. All three phases have different magnetic properties. The hcp and fcc phases are known to exist at room temperature, whereas the e phase is considered a metastable phase. The hcp phase of cobalt with anisotropic high magnetic coercivity is more useful for permanent magnetic applications, and the fcc cobalt NPs have soft magnetic properties. The use of polymers as substrates for microfluidic devices is relatively recent and has several decisive advantages. While microfluidic devices have been manufactured using several different types of polymers including poly (carbonate), poly (ester), poly (styrene), poly (ethylene terephthalate glycol), poly (methyl methacrylate), and poly(olefins), the most popular polymer material in the laboratories is the soft silicone elastomer, poly(dimethylsiloxane) (PDMS) (Makamba et al. 2003; McDonald and Whitesides 2002; Binyamin et al. 2003; Locascio et al. 2003; Xu et al. 2001; Susan et al. 2000a, b). PDMS is a hybrid polymer and its chemical structure is characterized by an inorganic Si–O backbone on which organic methyl side groups are attached. PDMS has a number of useful physical and chemical properties, it is optically transparent down to 280 nm offers good electrical resistivity, and adequate thermal conductivity (McDonald and Whitesides 2002). It is non-toxic, biologically inert and provides high permeability to gases. Concerning microfabrication, due to its low cost and easy handling, PDMS has become a primary material for the low-volume manufacturing of microfluidic devices. PDMS has extremely low glass transition temperature (-123 °C) giving it elastic properties at room temperature. Therefore, it deforms reversibly. This added to its low surface energy allows it to be peeled easily. PDMS also seals readily with glass and silicon oxide. PDMS however, suffers from some significant limitations. For example, pristine PDMS is hydrophobic in nature, so it can be difficult to wet and its surface is vulnerable to non specific adsorptions of molecules and

Microsyst Technol (2013) 19:409–418

surface charges. Finally, the inert surface provides no simple direct route for modification (Hu et al. 2002). PDMS is quite elastic in nature, thermally stable up to *250 °C and immobilizes the particles dispersed in it (Hu et al. 2002; Dow Corning Corporation 1991). Although PDMS can chemically protect the particles embedded in it, two issues can occur: (1) the high surface energy of NPs makes them susceptible to oxidation before their incorporation in the polymer matrix reducing their magnetic properties, (2) they may also agglomerate. Therefore, it is necessary to passivate the surface of cobalt particles, if possible during the synthesis itself. There are many reports on the synthesis of cobalt NPs (Yang et al. 2004; Shao et al. 2006; Petit et al. 1999; Grass and Stark 2006). Various surface passivation schemes using different capping molecules have been utilized. However, in most of the cases the synthesis temperature has been larger than *100 °C. In this work, we use a low temperature chemical route to synthesize cobalt NPs with proper capping to be incorporated in the PDMS matrix. The contribution of this work is in demonstrating a new magnetic nano-composite material based on the magnetic properties of cobalt and iron NPs with improved response to magnetic field compared to existing polymeric magnetic membranes. We first report on the chemical synthesis of passivated capped cobalt and water based cobalt/iron NPs. Then the fabrication of cobalt and iron NPs embedded PDMS membranes is detailed. Finally the NPs are characterized using various techniques and the magnetic and mechanical properties of the composite membranes are investigated. We also report the surface wettability of the cobalt nanocomposite PDMS under various plasma conditions (RF power, exposure duration and gas pressure) and compared it with bare PDMS, using contact angle measurement of water droplet and Fourier Transform Infrared Spectroscopy (FTIR).

2 Experimental 2.1 Chemical synthesis of magnetic nanoparticles 2.1.1 Water based uncapped cobalt and iron NPs Both cobalt and iron NPs were synthesized using the reduction of the metal ion with the most common reducing agent borohydride. The synthesis procedure used was through controlling the reduction of Co and Fe ions by sodium borohydride in aqueous solution at room temperature. Solution-1 was prepared by dissolving and sonicating 0.15 g of FeCl36H2O/CoCl26H2O (Molychem AR grade) in 50 ml of Millipore water. The other solution-2, also prepared by dissolving 0.4 g NaBH4 (Thomas Beaker

411

AR grade) in 20 ml of Millipore water was added dropwise (0.5 ml/min) in solution-1, the total transfer took about 60 min. The mixture was shaken by hand (magnetic stirring could not be used in order to avoid magnetically induced aggregation of the resultant magnetic NPs). Immediately after addition of solution-2, hydrogen gas bubbles appeared, and the mixture turned to a black colour slowly. It was left to settle for a few hours while the magnetic NPs grew, water was then carefully poured off the top of the mixture vessel and replaced by ethanol. A black precipitate was obtained, which was washed repeatedly five times with ethanol using a centrifuge. The NPs were then placed in a Petri dish and subjected to vacuum for 2–3 h to get the powder of Co/Fe NPs. 2.1.2 PEG/oleic acid capped Co NPs A two-step synthesis was carried out at room temperature by a chemical route in which cobalt NPs were produced first with an intermediate polyethylene glycol ((H–OCH2CH2)n–OH) (PEG) capping which was then replaced by an oleic acid (OA) capping to stabilize the NPs. The details of the synthesis procedure are given below. Two aqueous solutions were separately prepared, one dissolving 1 g CoCl26H2O in 50 ml de-ionized water and 10 ml of water soluble PEG 600. This first, pink coloured solution was constantly stirred for 15 min at 50 °C and then cooled down to room temperature, turning to dark red in colour. The second solution was prepared by dissolving 2 g of NaBH4 in 20 ml de-ionized water. This solution was colourless in appearance. The second solution was then slowly added to the first one under constant stirring, which changed the solution colour from dark red to black. Mixing of the two solutions was complete in about 45 min. This procedure produced PEG-capped cobalt particles, as verified by FTIR spectroscopy. It was observed that PEG capping is not stable, particles converting to their oxidized phase. This led to the addition of OA as a second, more stable capping molecule. In this second step, 2 ml OA were added drop wise in the above mixture under constant stirring, which was found to successfully completely replace the initial capping of PEG and produce chemically stable Co NPs. After settling down, the precipitate was repeatedly washed with ethanol to get the black powder of cobalt NPs. 2.2 Fabrication of magnetic nanoparticle-embedded PDMS membranes Magnetic composite PDMS membranes were fabricated by the casting method, based on the curing of PDMS in the liquid form. We used a two-part PDMS kit (Sylgard 184 from Dow Corning, MI, USA) consisting of a silicone prepolymer solution and a curing agent. Fabrication of the NPs

123

412

embedded PDMS membrane is a two-step process. In the first step, the PDMS prepolymer solution was mixed thoroughly with its catalyst at 10:1 ratio (weight:weight) and degassed in a desiccator under vacuum. An appropriate quantity of NP powder (by the weight percentage with respect to PDMS) was dispersed in chloroform to prepare the magnetic NPs solution. Chloroform was chosen to disperse the NPs as it reacts neither with PDMS nor with the OA capping. After mixing the NPs and PDMS solutions the mixture was kept in a vacuum desiccator to completely remove chloroform and any trapped air in the PDMS mixture. When the degassing was over, the PDMS membrane was cast, i.e., the mixture was filled in a Teflon mould and cured at 100 °C for 1 h. After cooling to room temperature, the freestanding membrane was carefully detached. PDMS composite membranes of various weight percentages of Co and Fe NPs (16 and 25 wt%) were fabricated. All the membranes were of about 50 mm diameter and *250 lm thickness. It was observed that, at higher weight percentage beyond 25 wt%, the weight of magnetic NPs was higher than that of PDMS solution, so it was rather difficult to form a homogenous and smooth membrane. 2.3 Characterization of nanoparticles and composite membranes Magnetic NPs and the membranes were characterized using various techniques. 2.3.1 Structural studies X-Ray Diffraction (XRD) studies of magnetic NPs powder were performed with an x-ray diffractometer (Bruker D8 ˚ ) radiation. Atomic Advance), using the CuKa (1.5406 A Force Microscopy (AFM) of the NPs was carried out in the contact mode with a silicon cantilever using a JEOL JSPM5200 scanning probe microscope. AFM samples were prepared by dispersing the magnetic NPs powder in chloroform and a thin film was formed on the Si (111) substrate. FTIR spectroscopy investigations of magnetic NPs and membranes were carried out at room temperature using a Thermo Nicolet 6700 spectrometer operated in the transmission mode.

Microsyst Technol (2013) 19:409–418

Fig. 1 Magnetic deflection measurement set-up

?7,000 Oe with a step size of 500 Oe/sec. The sensitivity of the measurement was 1 9 10-6 emu/g. An important characteristic of magnetic membranes is their deflection under the application of a magnetic field. The deflection of the membranes at the centre under the application of a magnetic field was measured using a simple lab-made set-up. As shown in Fig. 1 it consists of (1) a membrane holder capable of holding membranes of different sizes, (2) a magnet holder and (3) a travelling microscope. Positions of the membrane holder and the magnet holder can be adjusted along the vertical direction. A magnetic membrane under investigation can be placed in the membrane holder and the microscope is focussed at the centre of the membrane without applying any magnetic field. After this, the position of the membrane holder stays fixed throughout the measurements. The magnetic field can then be applied by placing a permanent magnet of 1 9 0.5 cm size on the magnet holder. The magnet was taken big enough for uniform magnetic field over the entire magnet. Presence of the magnet causes a deflection of the membrane depending upon the distance between the membrane and the magnet, which results into defocusing of the membrane image. The deflection is then measured by bringing the membrane back into the focus and noting the movement of the microscope with an accuracy of one micrometer. The magnetic field can be calibrated using a Gauss meter by placing it at the position of the membrane and moving the magnet position.

2.3.2 Magnetic studies

3 Results and discussion

The magnetic properties of the magnetic NPs and the membranes were investigated at room temperature using a Lake Shore 7304 Vibrating Sample Magnetometer (VSM). Samples were placed in a non-magnetic plastic tubular holder. The field was scanned in the range from -7,000 to

3.1 Uncapped magnetic cobalt and iron NPs

123

Uncapped magnetic NPs of cobalt and iron have been synthesized using the water based chemical route described above. These NPs were characterized by XRD and

Microsyst Technol (2013) 19:409–418

413

3.2 Capped cobalt NPs

Fig. 2 XRD pattern of as synthesized uncapped a cobalt and b iron nanoparticles

their XRD patterns are shown in Fig. 2. Figure 2 (a) shows the XRD pattern of cobalt NPs and its XRD analysis reveals a broadening of the peak at 44.15° which corresponds to the cobalt cubic phase. Other peaks are not clearly observed [it may be due to the more intense (111) peak]. The broad peak around 82° (420) is due to the e-cobalt phase. Figure 2b shows iron NPs XRD pattern. The diffraction peaks at (110), (200) and (211) planes of iron NPs are readily recognized and are consistent with a cubic structure. The average particle size of NPs was determined using the Scherrer formula, applicable in case of NPs d ¼ B0:9k cos h, where d is the particle size, k is the wavelength of the ˚ ), B is the full width at half maxix-rays used (1.5406 A mum of the diffraction peak expressed in radians and h is the angle of diffraction (18). It was found to be 28 and 30 nm for cobalt and iron NPs, respectively.

The synthesis of magnetic NPs like cobalt and iron can be challenging as they may quickly get oxidized and become non-magnetic in some cases or get aggregated and demagnetize as a whole. For example, the cobalt metallic particles are ferromagnetic while cobalt oxide is anti-ferromagnetic. Therefore, it is imperative to passivate the surface of the cobalt NPs with organic molecules to avoid their oxidation as well as their aggregation. Oleic acid is found to be a good capping agent for cobalt NPs due to its long carbon chain [C17H33COOH] (Shao et al. 2006; Petit et al. 1999). However, OA is immiscible in water and cannot be used in our aqueous synthesis procedure, where we need to use NaBH4 for the reduction of the cobalt salt. Therefore, we have adopted a two-step procedure in which we first carry out a synthesis with PEG ((H–OCH2CH2)n– OH) and then replace the PEG with OA. Although PEG coated iron NPs have been reported (Bonder et al. 2007) in our experience OA was much superior. In the synthesis procedure developed here we have used the maximum temperature *50 °C. The cobalt NPs, in the two step process, is initially capped with PEG 600 and then with OA. This was confirmed by FTIR technique. We obtained the cobalt NPs capped with OA in the powder form. This is evident from the FTIR spectra shown in Fig. 3. It shows the FTIR spectra of pure PEG-600 [spectrum (a)] and PEG-600 capped cobalt NPs [spectrum (b)]. The absorptions at 2,952 and 2,860 cm-1 in PEG spectrum are due to the CH2 stretching of aliphatic chains. These bonds convoluted to 2,978 cm-1 when PEG was attached to the NPs. Simultaneously a new bond at 1,642 cm-1 emerged, which may be attributed to a bonding state very similar to the CH2 stretching of aliphatic chains. The absorption around 1,092 is due to the C–O stretching (Mohan 2005). This peak is shifted to 1,079 cm-1 when PEG is attached to the NPs. These changes in FTIR spectrum clearly indicate the PEG-600 interaction with cobalt NPs. Hence we can conclude that cobalt NPs were initially capped with PEG-600 in the presence of water at room temperature. Further we replaced the PEG-600 capping with OA as OA is a good stabilizer for cobalt NPs. Figure 2c and d show the FTIR of OA capped cobalt NPs and pure OA, respectively. The strong absorptions in the region 2,912, 2,834 cm-1 and at 1,706 cm-1 in the finger print region are attributed to the aliphatic chain in the carboxylic acid group of pure OA. In case of OA capped NPs most of the peaks corresponding to OA absorption are present. The peaks corresponding to PEG are missing in this case. We observe that the 2,912 cm-1 peak in OA is shifted to 2,939 cm-1, whereas the 1,706 cm-1 peak is shifted to 1,712 cm-1, which indicates that the interaction of OA with the Co-NPs takes place via the aliphatic chain of OA.

123

414

Microsyst Technol (2013) 19:409–418

Fig. 4 XRD pattern of as synthesized oleic acid capped cobalt nanoparticles

Fig. 3 FTIR of a pure PEG-600 b PEG-600 capped cobalt nanoparticles c oleic acid capped cobalt nanoparticles d pure oleic acid

From the FTIR spectra, it can be concluded that OA is chemically bonded on the Co-NPs surface entirely replacing the PEG. This provides a very strong acidic and hydrophobic capping to the NPs providing stability against oxidation (Mohan 2005; Nianqiang et al. 2004). Crystallinity and size of the cobalt NPs have been determined using XRD. In Fig. 4 the XRD spectra have been illustrated. Diffraction peaks corresponding to the (111), (200) and (220) planes of the fcc phase are present in agreement with the JCPDS (Joint Committee for Powder Diffraction Standards) powder diffraction file (card # 15-0806) (JCPDS database 1999). The average particle size using the Scherrer formula is *40 nm. The particle size was also determined using AFM as shown in Fig. 5a with the size distribution plot in Fig. 5b. The average size of the particles is *37 ± 5 nm, which is quite close to that determined using XRD. Their small size is beneficial for uniformly embedding the particles in the polymer matrix. Magnetization measurements at room temperature (see Fig. 6; Table 1) of cobalt particles suggest that the particles are single domain particles with a magnetization value of 19.21 emu/g and a coercivity value of 51.49 Oe. The saturation magnetic moment for the bulk cobalt is reported to be 162.62 emu/g and the coercivity to be about 30 Oe (Pauthenet 1982; Keif et al. 1991). The size dependent reduction of the magnetic moment and the increase in the coercivity (which is a more complex function of size) are

123

observed in the case of NPs (Keif et al. 1991). The reduced magnetisation value can also be attributed to a possible magnetic dead layer at the surface or interface of the particles (Gajbhiye et al. 2008). Often the spins of the surface atoms are not able to respond to the applied magnetic field due to a pinning by the interface atoms, which results in a rise in the coercivity. Thus, the oleic acid capping of the Co NPs must be responsible for the increase in the coercivity and the reduction in the magnetization. 3.3 Magnetic membranes Magnetic NP-embedded (16 and 25 wt%) PDMS membranes were fabricated using uncapped and capped NPs with 50 mm diameter and *250 lm thickness. One of the main challenges in embedding NPs into a polymer matrix is the aggregation of particles which leads to non-uniform distributions of particles in the polymer matrix. Particles beyond 25 wt% in PDMS were difficult to mix properly, because the particle concentrations were high compared with the PDMS volume, making it difficult to achieve a homogeneous mixture. Deflection measurements were performed first with the uncapped cobalt and iron magnetic NPs embedded in the PDMS matrix using a 700 G magnet. Figure 7 shows a deflection plot of those membranes at different magnetic pressures. The higher the concentration of the magnetic NPs (cobalt and iron) in the PDMS matrix, the larger the deflection. Moreover, the membrane with 25 wt% cobalt shows a deflection (*500 lm) higher than that obtained for 25 wt% iron. Deflection studies were also performed for a capped cobalt NP-embedded membrane under the application of the magnetic field as is plotted in Fig. 8. The deflection obtained was *400 and *900 lm in case of the 16 and

Microsyst Technol (2013) 19:409–418

415

Fig. 5 a AFM image and b size distribution of cobalt nanoparticles

Fig. 6 Magnetization curves for a Co nanoparticles b a 25 wt% membrane, and c a 16 wt% membrane

Table 1 Comparison of magnetic data of cobalt particles and membranes Material

Magnetization (Ms, emu/g)

Coercivity (Hc, Oe)

Co bulk particle

162.62

30

Co particle (*40 nm)

19.21

51.49

Co membrane (25 wt%)

16.08

150.71

25 wt% membranes, respectively. These values are much larger than the earlier reported values for the uncapped cobalt and iron NP-based membranes studied in our work

and in the literature (iron and iron oxide NPs embedded in the PDMS matrix) (Berthier and Ricoul 2004). Further, we investigated the deflection of the magnetic membranes by subjecting them to magnetic fields in opposite directions, i.e., once the north pole of the permanent magnet faced the membrane and then the south pole faced the membrane. The field was varied by changing the distance from the magnet to the membrane in each case. The deflections caused in the membrane at different positions of the magnet were recorded in both directions, while the magnet approaching the membrane

123

416

Microsyst Technol (2013) 19:409–418

16wt% Co in PDMS 25wt% Co in PDMS

500

Deflection( µ m)

400

300

200

100

0 0

200

400

600

800

Fig. 8 Magnetic field versus deflection plot for OA-capped a 16 wt%, and b 25 wt% cobalt membranes

Magnetic Pressure (Pa) 15wt% Fe in PDMS 22wt% Fe in PDMS

80

Deflection( µ m)

60

40

20

0 0

200

400

600

800

Magnetic Pressure (Pa)

Fig. 7 Magnetic field versus deflection plot for uncapped 16 and 25 wt % a cobalt and b iron membranes

(increasing magnetic field) and while moving away from the membrane (decreasing magnetic field). In Fig. 9 typical data for the 16 wt% cobalt membrane is depicted. The extent of deflection varied in case of oppositely directed fields. This indicates that the interactions of the Co particles in the membranes are dependent on the magnetic field directions. Concerning the magnetic properties of these membranes using VSM, in case of the composite membrane with OA-capped cobalt NPs the magnetization value is lower than that of the OA-capped Co NPs, whereas the coercivity value increases (Table 1). This can be attributed to the presence of the PDMS matrix.

123

Fig. 9 Deflection of the 16 wt% Co-PDMS membrane under a forward and b reverse direction of magnetic field. The arrow indicates the increasing and decreasing direction of magnetic field

4 Conclusion and perspective The contribution of this work is in demonstrating a new magnetic polymeric membrane with improved response to the magnetic field compared to existing polymeric magnetic membranes. PDMS membranes embedding cobalt

Microsyst Technol (2013) 19:409–418

and iron NPs combine the favorable properties of magnetic nanomaterials with simple and economical processing of an elastomer to achieve large deflection. The main results of this work are summarized below: –

–

–

–

Water based cobalt and iron magnetic NPs have been synthesized using a simple modified chemical route in the absence of capping agent at room temperature. Oleic acid capped ferromagnetic cobalt NPs have been synthesized using a low temperature two-step chemical synthesis method. The cobalt NPs were of *37 nm in size and showed a single domain ferromagnetic behaviour. New freestanding magnetic membranes were fabricated by dispersing the uncapped (cobalt and iron NPs) and capped (cobalt) magnetic NPs with concentrations of 16 and 25 % by weight in the flexible silicone PDMS matrix using the casting method. The capped cobalt NP composite membranes had good flexibility up to *25 wt% loading. These membranes showed large and linear deflections under the application of magnetic field. The highest deflection was obtained with the 25 % OA-capped CoNP-PDMS membrane.

Magnetic flexible thin polymeric membranes have a great potential in polymer microsystems for actuation and sensing purposes. Polymer composite membranes can be easily integrated in a polymeric device, using the same microfabrication technologies as for a virgin polymer. The use of magnetic membranes in active devices would also be convenient as they can be actuated externally without the need of any direct connection or wiring, allowing for contact-less operation, which would not only ease operation, but could also bring down the cost of devices. In particular, we anticipate that the technology presented above can be advantageously applied to lab-on-a-chip applications where low cost and/or disposable aspects are of primary importance. For example, such membranes would be particularly useful in microfluidic systems where remote control of valves and micro mixers without any external wire connections to the device is desired in order to achieve appropriate flow and mixing of liquids. Acknowledgments This work was carried out within the frame work of CEFIPRA (IFCPAR project (# 3408-01)) and DST, India. A. Singh and M. Shirolkar would like to thank CEFIPRA for the financial support.

References Abyanch MK, Kulkarni SK (2008) Giant piezoreasistive response in zinc-polydimethylsiloxane composites. J Phys D Appl Phys 41:135405–135411

417 Abyanch MK, Paramanik D, Varma S, Gosavi SW, Kulkarni SK (2007) Formation of gold nanoparticles in polymethyacrylate be UV irradiation. J Phys D Appl Phys 40:3771–3779 Berthier J, Ricoul F (2004) Development of micromembranes as actuators in microsystems. Appl Nanosci 1:31–38 Binyamin G, Boone TD, Lackritz HS, Ricco AJ, Sassi AP, Williams SJ (2003) Plastic microfluidic devices: electrokinetic manipulations, life science applications, and production technologies. Lab on a chip, p 83–112 Bonder MJ, Zhang Y, Kiick KL, Papefthymion V, Hadjipanayis GC (2007) Controlling synthesis of Fe nanoparticles with polyethylene glycol. J Magn Mag Mater 311:658–664 Dow Corning Corporation (1991) SYLGARD 184, data sheet, Michigan Doyle PS, Bibette J, Bancaud A, Viovy JL (2002) Self-assembled magnetic matrices for DNA separation chips. Science 295:2237 Fahrni F, Prins FWJ, Van IJzendoorn LJ (2009) Magnetization and actuation of polymeric microstructures with magnetic nanoparticles for application in micro fluidics. J Magn Magn Mater 321:1843–1850 Gajbhiye NS, Sharma S, Nigam AK, Ningthoujam RS (2008) Tuning of a single to multidomain behaviour for monodispersed ferromagnetic cobalt nanoparticles. Chem Phys Lett 466:181– 185 Grass RN, Stark WJ (2006) Gas phase synthesis of fcc-cobalt nanoparticles. J Mater Chem 16:1825–1830 Gupta PK, Hung CT, Lam FC, Perrier DG (1988) Albumin microspheres III. Synthesis and characterization of microspheres containing adriamycin and magnetite. Int J Pharm 43:167–177 Hu S, Ren X, Bachman M, Sims CE, Li GP, Allbritton N (2002) Surface modification of poly (dimethylsiloxane) microfluidic devices by ultraviolet polymer grafting. Anal Chem 74:4117– 4123 JCPDS database (1999) Powder diffraction file PDF-2 Data Base International Center for Diffraction Data JCPDS-ICDD, Pennsylvania Jordan A, Scholz R, Wust P, Fahling H, Felix R (1999) Magnetic fluid hyperthermia (MFH): cancer treatment with AC magnetic field induced excitation of biocompatible superparamagnetic nanoparticles. J Magn Magn Mater 201:413–419 Keif MT, Mankey GJ, Willis RF (1991) Micromagnetic properties of ultrathin cobalt films. J Appl Phys 69:5000–5002 Khoo M, Liu C (2001) Preparing of a high speed bistable electromagnetic RF MEMS switch. Sens Actuators A 89:259–266 Locascio LE, Henry AC, Johnson TJ, Ross D (2003) Surface chemistry in polymer microfluidic systems. Lab on a chip, p 65–82 Makamba H, Kim JH, Lim K, Park N, Hahn JH (2003) Surface modification of poly (dimethylsiloxane) microchannels. Electrophoresis 24:3607–3619 McDonald JC, Whitesides GM (2002) Poly (dimethylsiloxane) as a material for fabrication of microfluidic devices. Acc Chem Res 35:491–499 Mohan J (2005) Organic spectroscopy principles and applications. 2nd edn. Narosa Publishing House, New Delhi Nianqiang W, Fu L, Su M, Aslam M, Wong KC, Dravid VP (2004) Interaction of fatty acid monolayers with cobalt nanoparticles. Nano Lett 4:383–386 Oswald P, Clement O, Chambon C, Schouman-Claeys E, Frija G (1995) Liver positive enhancement after injection of superparamagnetic nanoparticles: respective role of circulating and uptaken particles. Magn Reson Imaging 15:1025–1031 Pan T, Kai E, Matthew S, Barocas V, Ziaie B (2004) A magnetically driven PDMS peristaltic micropump. Proceedings of 26th Annual International Conference of the IEEE EMBS, San Francisco, pp2639–2642

123

418 Pan T, McDonald SJ, Kai EM, Ziaie B (2005) A magnetically driven PDMS micropump with ball check-valves. J Micromech Microeng 15:1021–1026 Park IW, Yoon M, Kim YM, Kim Y, Kim JH, Kim S, Volkov V (2004) Synthesis of cobalt nanoparticles in polymeric membrane and their magnetic anisotropy. J Magn Magn Mater 272–276: 1413–1414 Pauthenet R (1982) Experimental verification of spin-wave theory in high fields. J Appl Phys 53:8187–8192 Perez JM, Simeone FJ, Saeki Y, Josephson L, Weissleder R (2003) Viral-induced self-assembly of magnetic nanoparticles allows the detection of viral particles in biological media. J Am Chem Soc 125:10192–10193 Petit C, Taleb A, Pileni MP (1999) Cobalt nanosized particles organized in a 2D superlattice: synthesis, characterization, and magnetic Properties. J Phys Chem B 103:1805–1810 Pirmoradi F, Cheng L, Chiao M (2010) A magnetic poly (dimethylesiloxane) composite membrane incorporated with uniformly dispersed, coated iron oxide nanoparticles. J Micromech Microeng 20(015032):1–7 Rosensweig RE (1985) Ferrohydrodynamics. Cambridge University Press, Cambridge Shao H, Huang Y, Lee H, Suh YJ, Kim CK (2006) Effect of surfactants on the size and shape of cobalt nanoparticles synthesized by thermal decomposition. J Appl Phys 99:702–705 Speliotis DE (1999) Magnetic recording beyond the first 100 years. J Magn Magn Mater 193:29–35

123

Microsyst Technol (2013) 19:409–418 Susan LR, Barker Ross D, Tarlov MJ, Gaitan M, Locascio LE (2000a) Control of flow direction in microfluidic devices with polyelectrolyte multilayers. Anal Chem 72:5925–5929 Susan LR, Barker Tarlov MJ, Canavan H, Hickman JJ, Locascio LE (2000b) Plastic microfluidic devices modified with polyelectrolyte multilayers. Anal Chem 72:4899–4903 Wilson JL, Poddar P, Frey NA, Srikanth H, Mohomed K, Harmon JP, Kotha S, Wachsmuth F (2004) Synthesis and magnetic properties of polymer nanocomposites with embedded iron nanoparticles. J Appl Phys 95:1439–1443 Xu W, Uchiyama K, Shimosaka T, Hobo TJ (2001) Fabrication of polyester microchannels and their applications to capillary electrophoresis. Chromatogr A 907:279–289 Yamahata C, Lacharme F, Gijs MAM (2005) Glass valveless micropump using electromagnetic actuation. Microelectron Eng 78–79:132–137 Yang HT, Shen CM, Wang YG, Su YK, Yang TZ, Gao HAT (2004) Stable cobalt nanoparticles passivated with oleic acid and triphenylphosphine. Nanotechnology 15:70–74 Yang YM, Tsai HH, Hsu TH, Su YC (2009) Magnetic shaping of ferrite-PDMS composite and its application in magnetic carrier targeting. Transducers 1983–1986 Yufeng S, Wenyuan C, Feng C, Wieping Z (2006) Electro-magnetically actuated valveless micropump with two flexible diaphragms. Int J Adv Manuf Technol 30:215–220