Microcontact printing of polyelectrolyte multilayer thin films: Glass–viscous flow transition based effects and hydration methods

Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 271–278

Contents lists available at ScienceDirect

Colloids and Surfaces A: Physicochemical and Engineering Aspects journal homepage: www.elsevier.com/locate/colsurfa

Microcontact printing of polyelectrolyte multilayer thin films: Glass–viscous flow transition based effects and hydration methods Meiyu Gai a,b , Johannes Frueh a,∗ , Gleb B. Sukhorukov b , Agnes Girard-Egrot c , Samuel Rebaud c , Bastien Doumeche c , Qiang He a,∗ a Key Laboratory of Microsystems and Microstructures Manufacturing, Ministry of Education, Micro/Nano Technology Research Centre, Harbin Institute of Technology, Yikuang Street 2, Harbin 150080, China b Queen Mary University of London, Mile End, Eng, 215, London E1 4NS, United Kingdom c Institut de Chimie et Biochimie Moléculaires et Supramoléculaires, Universite Claude Bernard Lyon 1, 43 boulevard du 11 Novembre 1918 F-69622 Villeurbanne cedex, France

h i g h l i g h t s

g r a p h i c a l

• Glass–viscous flow transition effects

Micro and nanostructured surfaces and samples are of fundamental importance for electronics, tissue engineering and drug delivery. The effect of glass–viscous flow transition on thin polymer films for microcontact printing was investigated by the example of polyelectrolyte multilayers depending on the softening method (cold versus hot solvent) and for different temperatures in relation to the glass–viscous flow transition point. Interestingly PEM structures can not only be printed but also be expelled from stamps in aqueous solution when the stamp is removed. This is due to emerging osmotic pressures created by dissolving PEM at temperatures exceeding the PEM glass transition point.

• • • •

influences the quality of PEM patterns during printing. PEM that dissolves within stamp structure increases osmotic pressure. Osmotic pressure can be used to expel printed PEM structures from stamps. Cold water vapour from a humidifier can be used for humidifying PEM. Active printing area necessary for printing PEM not total stamp area.

a r t i c l e

i n f o

Article history: Received 28 March 2015 Received in revised form 4 May 2015 Accepted 11 May 2015 Available online 20 May 2015 Keywords: Polyelectrolyte multilayers Microcontact printing Capillary forces Osmotic pressure PEM glass–viscous flow transition Humidification method

a b s t r a c t Micro and nano-patterned surfaces are important for many applications ranging from antibiofouling over tissue engineering to electronics. Often the incorporation of functional entities is of interest. Polymer coatings especially polyelectrolyte multilayer (PEM) films and patterns are materials offering a large variety of tuning and engineering. The PEM pattern printing quality bases not only on the surface force balance but also in the way the PEM is softened, which can be done by printing the PEM in water, using an ultrasound humidifier or by exposing the film to (hot) water vapor. In this publication it is shown, that cold water vapor from an ultrasound humidifier or direct printing in water is superior to steam evaporation onto PEM thin films as humidification method. In addition the capillary pressure of the patterns within the stamp and the glass–viscous flow transition point of the PEM thin film are the significant parameters for PEM printing. This is because the PEM can surpass the glass–viscous flow transition point due to the shear forces and be sucked into the stamp microwells (or holes) preventing a structure replication. Under high temperatures and in aqueous conditions, the PEM can be expelled from the microwells due

∗ Corresponding authors. Tel.: +86 15804610642. E-mail addresses: [email protected] (J. Frueh), [email protected] (Q. He). http://dx.doi.org/10.1016/j.colsurfa.2015.05.009 0927-7757/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t

272

M. Gai et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 271–278

to the osmotic pressure produced by the counter ions of PEM in glass–viscous flow state and dissolving polyelectrolyte if a PEM with counter ion based charge balance is used. © 2015 Elsevier B.V. All rights reserved.

1. Introduction

2. Experimental

The creation of micro and sub-microstructures is of fundamental interest for numerous applications ranging from biosensors [1] over electronic [2] to antifouling [3,4] surfaces. Out of a large variety of potential techniques micro and sub-microstructures can be conveniently produced by microcontact printing (␮CP) [5]. Within the last 20 years thin polymer films produced by the layer-by-layer [6] (LbL) bottom up approach gained significant interest. These polymer films comprise out of electrostatically charged polymers called polyelectrolyte (PE) and are assembled in an alternating manner to a multilayer (PEM). The PEM thickness can be controlled down to sub-nanometer precision and even contain living cells and other functional entities like nanoparticles [7,8]. For the design of real or reproduced tissue or drug delivery systems [9] it is of fundamental importance to control the PEM structure and shape. One method is the use of templates [9], another one was introduced by Hammond and Park in the year 2004 by simply ␮CP PEM thin films using a polydimethylsiloxane (PDMS) stamp (silicone rubber stamp). Since then the system was used by a variety of researchers [10–13] from different groups including our own, which also tried to explain the effects responsible for ␮CP. Since a lot of research has been done to explain the PEM printing effects, we summarize the current state of the art of ␮CP on a small overview of existing explanations and problems in the supporting information (SI). Briefly, ␮CP of a solid PEM film printed at low pressures is based on surface forces and PEM mechanical properties, while high pressure can cause a liquefaction of the PEMs [11,13–15]. The printing limit of PEM is PEM surface energy and line tension and not PDMS structure based [5,13]. In case applying high pressure the detailed effects are not fully understood especially for systems containing PEM in viscous flow state, where reports are very scarce, no temperature or different pressures and PEM composition are available. Park et al. was to the knowledge of the authors the first to investigate the effect of capillary pressure on the printing quality, while the Shen group reported PEM flowing effects at high pressure printed PEM [11,16]. It is worth to point out that not only PDMS stamps were utilized for low pressure condensed PEM printing but some groups used a dissolvable PMMA stamp instead of PDMS stamps [17–19]. In this case the consideration of the relevant surface forces is not necessary because the PMMA can be dissolved in organic solvent releasing the PEM patterns. Other groups, on the contrary, focused on transfer methods, without the need to dissolve the PMMA stamps [20]. During the printing process the osmotic and capillary forces are similar for PMMA and PDMS based stamps and just the force strengths differ. Therefore the results presented in this publication are extendable to PMMA stamp based systems. In this report, we study the printing of PEM films with short printing times in the range of seconds, and pressures up to 50 g/cm2 , whereby the focus lies on the glass–viscous flow transition effects of PEM films and how to prevent PEM from undergoing glass–viscous flow transition. We also discuss effects of the printing time, which can also affect the flow of PEM in viscous flow state, an effect ignored in other publications trying to print liquefied PEM systems [11].

2.1. Production of PEM thin films The utilized PEM thin films were produced by spraying deposition method [21] onto flexible silicone rubber sheets (PDMS) (Dow Corning Midland, USA) which was produced by mixing component A and B in 10:1 ratio, degassing it for 30 min in vacuum and curing it for 2 h at 70 ◦ C. The air–PDMS interface was used for PEM assembly. The spraying time for each solution including rinsing water was 6 s with a distance between spraying bottle and sample of 15 cm. The used water was ultrapure water (18.2 MOhm cm, Elga Labwater, Beijing, China). The spraying cans were DC (Duennschicht Chromatographie) Spruehflaschen (type Air Boy, Nr. 0110.1) purchased from Carl Roth, Germany. The used polyelectrolytes (PE) were polyacrylic acid (PAA) with a molecular weight (MW) of 1800 g/mol, polystyrenesulphonate (PSS, MW 70,000 g/mol), poly(diallyldimethylammonium chloride) (PDDA, MW 100,000-200,000 g/mol), polyallylamine hydrochloride (PAH, MW 56,000 g/mol), polyethylenimine (PEI, MW ∼750,000 g/mol). The FITC (fluoresceine isothiocyanate), which was chemically linked to PAH, according to ref. [22] was along with the PE purchased from Sigma, St. Luis, USA. The PE concentration for all PE solutions was 0.5 g/L with an ionic strength of 0.5 Mol/L (NaCl, Chemical Reagents, Tianjin, China). The PE solution spraying sequence, which determines the PEM build-up structure, was: 1. 2. 3. 4.

PEI(PSS-PDDA)4 (PSS-PAHFITC)2 ; PEI(PSS-PDDA)4 (PSS-PAHFITC)2 PSS; PEI(PAA-PAH)4 (PAA-PAHFITC)2 ; PEI(PAA-PAH)4 (PAA-PAHFITC)2 PAA.

Whereby the terminal layers of samples 1 and 3 are positively and samples 2 and 4 are negatively charged. This allows a defined comparison of the effect of internal charge ratio and surface charge on the PEM printing properties at different temperatures. Printing pressures of 0, 10, 20, and 50 g/cm2 were used. The investigated temperatures were 22, 25, 30, 34 and (for preliminary studies) 37 ◦ C with a printing time of 5 or 45 s. 2.2. PEM softening methods Three different PEM softening methods were compared: 1. After the PEM was produced it was dried with N2 first, then humidified with water vapor produced from water heated to 65 ◦ C for 5 s (distance 1 cm), then the stamp was pressed onto the PEM film with above parameters, then removed. 2. After the PEM was produced it was dried with N2 then humidified with water vapor from an ultrasound humidifier (Yadu SC-M20, Henan Yadu Industrial Co. Ltd., Henan, China) 1 cm distance, 5 s evaporation time, power wheel adjusted to a position which is ∼10◦ above off. Afterwards the stamp was pressed onto it with the above stated parameters, and then removed.

M. Gai et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 271–278

273

3. The stamp was pressed onto the PEM directly after production where it was still in wet condition for the above stated parameters, and then the stamp was removed. For silicon stamps and investigation methods please see SI page 4. 2.3. Theoretical section Well known Eqs. (1) and (2) were used for the determination of the capillary pressure and the osmotic pressure [23]. pc =

2 cos  r

(1)

pO =

−RT In (fX) V

(2)

In Eq. (1) r is the radius of the microwells (holes) in the silicon master (2.5, 4 and 5 ␮m) (if poly(methyl methacrylate) (PMMA) or other stamps are used, the stamp structure in these needs to be used),  the surface energy (silicon dioxide 143.1 mN/m2 ) and  is the contact angle of water on silicon dioxide (∼15◦ ). In Eq. (2) R is the ideal gas constant, T the absolute temperature of the experiment, V the molar volume (m3 /mol) of the PE monomer units and their respective counter ions, f the activity coefficient (assumed 1) and  is the molar fraction of solvent (here water). It is pointed out, that the values for the osmotic pressure could only be determined indirectly, since the PEM thickness is very small. The minimum osmotic pressures were estimated from expelled PEM plates out of the holes, as well as from the fact, that after drying the printing water PEM “sludge” was found. For this reason a minimum PEM and counter ion dissolution degree could be estimated. 3. Results 3.1. Investigation of the produced silicon stamps As shown in SI Figure S1, the created silicon stamps have regular microwells and a flat outer surface. The inside of the wells exhibit regular and repeatable structures in every microwells covering ∼30% of the well volume. For this reason a flat PDMS sheet covered with PEM was used and the silicon stamp was pressed on the PDMS sheet, intending to lift off the PEM. In contrast to normal PDMS stamps this method also protects the silicon masters from PDMS contaminations. 3.2. Forces related to printing (1) Capillary pressure: Using our equation for the capillary pressure we get for our structures values of 55.8 mega Pascal (MPa) in the 5 ␮m and 27.9 MPa in the 10 ␮m microwells. These values are high enough to suck in solution and PEM in viscous flow state. (2) Osmotic pressure: To take this effect into account we compared the glass–viscous flow transition point of PEM, with the resulting expelled PEM plates, the resulting patterns and the calculated capillary forces. To understand this approach it is important to know that PEM surpassing the glass transition temperature can dissolve, if the PEM contains an intrinsic charge mismatch that is balanced by counter ions. The data of self-dissolving PEM in viscous flow state was obtained from a paper published by Koehler, showing that non-crosslinked PEM films can, when heated in solution, undergo a glass–viscous flow transition and even self-repel and rupture due to an intrinsic charge mismatch [24–26]. Since our used PEM films are similar to the system of Koehler, the idea is feasible (and proven by Park [11]), that this type of PEM film, when printed in presence of water, will undergo a glass–viscous

Fig. 1. Capillary pressure for 5 and 10 ␮m diameter conical microwells (1 and 2); osmotic pressure for the case that PEM below a well dissolves completely (3); osmotic pressure for a case in which 3× of the PEM below a microwell which flows into it due to capillary pressure and then dissolves completely (4); osmotic pressure of a 1 mol/L NaCl solution shown for comparison reasons (5).

flow transition due to mechanical stress [27,28]. This PEM in viscous flow state is then easily sucked into the holes of the stamp by capillary forces, leading to an increase in plate thickness. While the PEM is at low temperatures in a glassy frozen state (kinetically entrapped), exceeding the glass transition causes a high osmotic pressure due to counter ion and PE release of the PEM into solution [29]. This release of ions into solution is until now mainly observed for positively charge terminated PEM, while theoretically also possible for negatively charge terminated PEM, no reports to the knowledge of the authors are present [24,26,30,31]. This release of counter ions is slower than the suction into the microwell for example in publication [31] heating in time scale of minutes was necessary to completely dissolve PEM capsules. Therefore the PEM only dissolves within the pore where it causes a gradual increase in osmotic pressure, which balances and finally overcomes the capillary pressure. Upon release of the printing stamp and subsequent flushing with water the osmotic pressure outside of the hole is lower than inside, thus releasing the PEM plates as shown in Scheme 1. A theoretical calculation of the maximum osmotic pressure shows, that the complete dissolution of a thin film below a cylindrical microwell would not be enough to balance the capillary pressure. In fact, the utilized silicon stamp microwell is on the inside not of cylinder but rather of conical shape (SI Figure S1 and Fig. 1), which raises the maximum possible osmotic pressure above the capillary pressure. The fact that the capillary pressure sucks PEM in viscous flow state and water into the capillary, before it can dissolve, raises the osmotic pressure further (Fig. 2). The flow of up to 3× the pristine PEM amount was assumed, whereby experiments showed, that these values can be easily surpassed under the right conditions (up to factor 12, which is easily possible because the microwell area is just 5% of the silicon stamp, allowing an theoretical out of plate PEM excess of factor 20). 3.3. Effect of counter ions It is pointed out, that this expulsion method is only valid for the samples with a positive charge termination produced in high ionic strength solvent, since this type of sample is self-dissolving due to an excess of counter ions which shield an excess of positively charged PE [24,32,33]. This PE based charge mismatch is caused by the PE charges being shielded by counter ions [24,32,33]. The PE coils in solution and is embedding counter ions that way

274

M. Gai et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 271–278

Scheme 1. PEM micro-plate creation in aqueous environment by utilizing a mixture of capillary force and osmotic pressure: (a) sample structure before printing; (b) PEM glass–viscous flow transition outside of the microwells due to shear forces and pressure and flow into the microwell due to capillary forces; (c) buildup of osmotic pressure due to surpassing the glass transition point and PEM dissolution; (d) expulsion of the PEM micro-plates upon lift off of the patterned silicon stamp due to osmotic pressure.

[24,32,33]. The degree of charge shielding and ion embedding of the PE defines the degree of entrapped ions [24,32,33]. The adsorption of oppositely charged PE releases these counter ions causing a charge balance [24,32,33]. In case of counter ion balanced charge, increasing the temperature leads in low ionic strength solvent to a counter ion release causing the positively charge terminated PE to self-repel and be released as well [26]. Negatively charged PEM like in case of sample types 2 and 4 is also sucked into the capillaries but resides therein and is not expelled, because the charges are shielded by PE and the counter ion amount is too low [24]. The counter ion release is the main driving force for the osmotic pressure. In our calculation the release of counter ions influences the osmotic pressure by factor ∼10 more than the PE release. This finding is in line with results of others that reported the entropy of counter ion release being the main driving force upon PEM formation [24,30]. Please note, that the PEM thickness of 30–60 nm is small compared to the well height and therefore conical structures in the range of microns are not significant. 3.4. Temperature effects for positively charge terminated PEM films For sample type 1 at 22 ◦ C and 50 g/cm2 pressure a success rate of ∼5% was achieved for direct printing. It was possible to decrease the pressure due to the glass–viscous flow transition based decrease of the line tension [13] at 32 ◦ C down to 10 g/cm2 for a success rate of 5%. At other temperatures and pressures the success rate for direct printing was close to 0, while on the contrary the expelled PEM plates as well as random PEM residues in viscous flow state increased with increasing temperature and printing time, as Fig. 4 shows. At high pressure, and high temperatures the failure of the direct printing method is due to PEM glass–viscous flow transition based effects, as large amounts of PEM in viscous flow states show (see SI Figure S2). This is shown most clearly when the temperature approaches or surpasses the glass–viscous flow transition temperature of the PEM, as shown in Fig. 4d. For sample type 3 the success rate of the direct printing method is for all observed temperatures (22–32 ◦ C) much higher (50–95% (depending on applied pressure; 50% for 0 and 10 g/cm2 , 95% for 20 and 50 g/cm2 )). This is because the glass–viscous flow transition temperature of this type of PEM is much higher than for PSS-PDDA based films, which has its glass–viscous flow transition

temperature at 37 ◦ C, while PAH based films experience this transition in the temperature range of 121–138 ◦ C as determined in autoclave measurements [34]. For this reason the amount of determined PEM in viscous flow state, as well as expelled plates is close to 0 (1–5 plates per printing attempt). 3.5. Temperature effects for negatively charge terminated PEM films of samples 2 and 4 No temperature but a dry or wet printing based effect was found, whereby the PEM stuck in the holes for dry printing, but for the wet printing approach the PEM was smeared and in viscous flow state on the patterned silicon stamp. For sample 4 the amount of PEM in viscous flow state was found to be less than for sample 2. 3.6. Properties of expelled positively charged PEM plates It is noted, that within the remaining water from the printing of spherical PEM micro-plates for sample type 1 on the PDMS substrate many dispersed micro-plates were swimming. These showed a much higher fluorescence compared to the PEM film (factor 10–12 higher), and were also exhibiting a better structural integrity, compared to the printed PEM micro-plates on the PDMS substrate. Some micro-plates exhibited heights, and even structural defects which were in the range and structure of the silicon stamp that was pressed onto the PEM. Due to the fact, that micro-plates were directly released into solution, they could not directly be counted. The random adsorption of these plates on the PDMS substrate also made a quantitative evaluation difficult, since a lot of plates were scattered around the substrate. Therefore a fraction of the sample was investigated, the plates counted and then extrapolated for the whole printed area. The efficiency of such a type of micro-plate release was for the wet printing process ∼0.01–0.5% (22–34 ◦ C) (samples 3 and 1) with preliminary experiments performed with sample 1 at 34–37 ◦ C reaching ∼15%. This effect shows the nonlinear increase in efficiency when the temperature reaches the glass transition point. The PEM plates remaining within the silicon master were below 5% (of all produced plates of sample type 1). In water which was flown from the PDMS substrate to the glass slide and dried there, clear signs of dissolved PAH-FITC were found, along with pieces of lifted off and partly dissolved PEM (probably due to shear force caused

M. Gai et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 271–278

275

Fig. 2. Printed PEM structures for sample 1 (a) fluorescence image of plates printed on PDMS with low pressure and temperature (5 ␮m silicon stamp, printed at 22 ◦ C, 50 g/cm2 pressure); (b) fluorescence image of plates expelled from the silicon master due to osmotic pressure effect (5–8 ␮m diameter, 5 ␮m silicon stamp, printed at 37 ◦ C, 50 g/cm2 pressure); Fluorescence image and its intensity section (c) (∼12× plate intensity compared to underlying PEM, same sample like in b)); SEM image of a plate expelled by the osmotic pressure approach (d) (same sample like b); AFM image of a plate expelled by the osmotic pressure approach (e) with section cut (f) and volume close to theoretical limit (20× film volume outside plate) and a structure similar like SI Figure S1 c) (same sample like in b)).

by the manual printing procedure, see Fig. 3, proving the pressure based PEM glass–viscous flow transition and dissolving effect). Please note at this point, that the properties of PEM in viscous flow state is different from dissolved PEM and the properties of PEM in viscous flow state are due to low amounts of available publications [25,27,31,35] still not fully understood. An interesting observation is that weights placed onto the middle of silicon stamps and decreasing the printing time led to a decrease of the direct expulsion effect compared to using manual pressure by hand. This observation proves the importance of an even printing pressure and

decreasing shear force, as well as shear force induced glass–viscous flow transition of PEM. The printing temperature is essential for this effect as well, since the printing pressure as well as shear pressure (the plates themselves do not experience pressure, but due to elongation of the PDMS, the PEM of the plates also undergoes a glass–viscous flow transition [27]) lead to a small increase in temperature. If this increase in temperature is not enough to overcome the glass–viscous flow transition temperature of the PEM, the printing process will follow the printing mechanism of ref. [13].

276

M. Gai et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 271–278

Fig. 3. Fluorescence microscopic images of (a) Sample 1 (22–25 ◦ C and 10–50 g/cm2 pressure success rate: 5% good print but many imperfections); (b) sample 1 for 32 ◦ C, even pressure, the amount of expelled PEM plates increases slightly, as does (c) PEM in viscous flow state, whereas direct printed pattern decrease; (d) 37 ◦ C, sample 1, uneven pressure, many expelled plates according to the capillary and osmotic pressure theory.

At higher pressures and less pattern density as well as higher temperatures than those applied in this study, the effect of capillary and osmotic pressure might produce a higher efficiency, which might lead to a system like the one of Park [11] or to a combination of the one of Park [11] and the osmotic pressure effect observed here. Such high pressures were however not in the interest of the proposed system.

from non-aqueous solvents [36–38]. As reported earlier [13] the patterned PEM made by a silicon stamp in cold aqueous conditions leads to more reliable patterns, since drying times and drying effects can be ignored. It was additionally found that cold water vapor from a humidifier not only causes more reliable structures than hot water vapor, but the drying times are also slower, offering longer printing and handling times. The printing in water was found to cause similar results like cold water vapor.

3.7. PEM softening method 4. Discussion Earlier references recommended a method for transferring PEM micro-plates from a PDMS stamp to a glass slide coated with PVA (polyvinylalcohol) (thin film, produced by spin-coating a saturated solution onto a silicon wafer at 4000 rpm) [15]. These methods utilized vapor of 65 ◦ C heated water to soften PEM micro-plates before transferring them to PVA [15]. Since the condensation enthalpy of water is very high, the hot water steam significantly heats PEM thin films and liquefies those which have a low glass–viscous flow transition point [25] and therefore these films experience a loss of structural integrity. Additionally hot water steam can heat and soften sacrificial layers like PVA, while the PEM is in viscous flow state at the same time, causing additional loss of structures. This method was compared with the results of printing with cold water vapor created by an ultrasound humidifier versus printing in water. Comparing cold versus hot water vapor shows that the printed structures on a PVA substrate are significantly better for cold water vapor, compared to the hot water vapor from hot water. In Fig. 4 it can be seen, that the high temperature led to a partial glass–viscous flow transitioning of the thin film, which then closed the spacing between the printed layers. Another problem that can be avoided with cold water vapor is that sacrificial layers like PVA start to dissolve, which prevents subsequent modification possibilities of the PEM like, e.g. PE adsorption

The capillary based theory presented here, with additional considerations of the glass–viscous flow transition properties of PEM are, together with the surface and line tension based theory published recently [13] able to explain all results obtained for PEM printing known to the authors. The fact, that in early systems like those of Park and Hammond [10,12] mainly thin PEM films were printed, and the PSS-PDDA films utilized a low ionic strength is showing that the PEM needs: (1) A strong surface charge/energy, to overcome the line tension (high salt concentration reduces the PE charge density [29]). (2) The PEM should not melt (the glass–viscous flow transition point of PEM depends also on the ionic strength) [25,26], when a clear pattern is desired and no direct expulsion or phase printing is intended. (3) The PEM must be soft enough to overcome the line tension; therefore, the printing was performed at 100% relative humidity or even in water. For a detailed comparison of the results presented here with existing systems and theories please see supporting information page 5–8.

M. Gai et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 271–278

277

Fig. 4. Fluorescence microscopic images comparing the printing quality from PDMS substrate to PVA using (a) cold vapor from an ultrasound humidifier with (b) hot water vapor from hot water. The integrity of the printed 25 ␮m strips printed on PVA is much better for cold water vapor. Printed structure is in both cases (PAA-PAH)3 (PSSPDDA)10 (PAHFITC-PSS)2 PAH. The glass–viscous flow transition point of such a film is ∼36 ◦ C.

5. Conclusion The presented possibility of using cold water from an ultrasound humidifier, allows the PEM printing to be more repeatable and faster and offers an interesting alternative for wet printing procedures, especially when water based systems are not intended or a dry production process is needed. Additionally the possibility of direct expulsion of PEM structures into solution without the need of sacrificial substrates was demonstrated. Although increased temperature is needed and longer printing times might be necessary, this effect holds promising potential in case of high pressure or heated silicon stamps. The finding that the absolute printing pressure differs depending on the real contact area allows a better estimation of the printing pressure needed for achieving repeatable structures or estimate the possibilities for glass–viscous flow transition effects. The proposed system was not only tested with synthetic polymers but was found to work also well with biomolecules like poly-l-lysine and chitosan.

[9]

[10]

[11]

[12]

[13]

[14]

[15]

Acknowledgment The authors would like to thank the start-up grant of HIT, China Postdoctoral Science Foundation (2013M531019), China Scholarship Council (201406120038) and EU-FP7 BIOMIMEM staff exchange project for funding this research.

[16]

[17]

[18]

Appendix A. Supplementary data [19]

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.colsurfa.2015.05. 009

[20]

References

[21]

[1] Z. Siwy, L. Trofin, P. Kohli, L. a Baker, C. Trautmann, C.R. Martin, Protein biosensors based on biofunctionalized conical gold nanotubes, J. Am. Chem. Soc. 127 (2005) 5000–5001, http://dx.doi.org/10.1021/ja043910f [2] G.I. Mijares, D.R. Reyes, J. Geist, M. Gaitan, B.J. Polk, D.L. DeVoe, Polyelectrolyte multilayer-treated electrodes for real-time electronic sensing of cell proliferation, J. Res. Natl. Inst. Stand. Technol. 115 (2010) 61, http://dx.doi.org/10.6028/ jres.115.005 [3] S.Y. Wong, L. Han, K. Timachova, J. Veselinovic, M.N. Hyder, C. Ortiz, et al., Drastically lowered protein adsorption on microbicidal hydrophobic/hydrophilic polyelectrolyte multilayers, Biomacromolecules 13 (2012) 719–726, http://dx. doi.org/10.1021/bm201637e [4] J. Frueh, M. Gai, Z. Yang, Q. He, Influence of polyelectrolyte multilayer coating on the degree and type of biofouling in freshwater environment, J. Nanosci. Nanotechnol. 14 (2014) 4341–4350, http://dx.doi.org/10.1166/jnn.2013.8226 [5] H.A. Biebuyck, N.B. Larsen, E. Delamarche, B. Miche, Lithography beyond ight: microcontact printing with monolayer resists, IBM J. Res. Dev. 41 (1997) 159–170, http://dx.doi.org/10.1147/rd.411.0159 [6] G. Decher, J.D. Hong, J. Schmitt, Build up of ultrathin multilayer films by a self-assembly process III. Consecutively alternating adsorption of anionic and cationic polyelectrolytes on charged surfaces, Thin Solid Films 210–211 (1992) 831–835. [7] A.J. Swiston, C. Cheng, S.H. Um, D.J. Irvine, R.E. Cohen, M.F. Rubner, Surface functionalization of living cells with multilayer patches, Nano Lett. 8 (2008) 4446–4453, http://dx.doi.org/10.1021/nl802404h [8] D.A. Gorin, S.A. Portnov, O.A. Inozemtseva, Z. Luklinska, A.M. Yashchenok, A.M. Pavlov, et al., Magnetic/gold nanoparticle functionalized biocompatible

[22]

[23] [24]

[25]

[26]

[27]

[28]

[29]

microcapsules with sensitivity to laser irradiation, PCCP 10 (2008) 6899–6905, http://dx.doi.org/10.1039/b809696a E. Donath, G.B. Sukhorukov, F. Caruso, S.A. Davis, H. Möhwald, hollow polymer Novel shells by colloid-templated assembly of polyelectrolytes, Angew. Chem. Int. Ed. Engl. 37 (1998) DOI http://onlinelibrary.wiley.com/doi/10.1002/(SICI)15212201–2205, 3773(19980904)37:16<2201::AID-ANIE2201>3.0.CO;2-E/abstract. J. Park, P.T. Hammond, Multilayer transfer printing for polyelectrolyte multilayer patterning: direct transfer of layer-by-layer assembled micropatterned thin films, Adv. Mater. 16 (2004) 520–525, http://dx.doi.org/10.1002/adma. 200306181 J.S. Park, S.M. Cho, G.Y. Han, S.J. Sim, J. Park, P.J. Yoo, Phase controllable transfer printing of patterned polyelectrolyte multilayers, Langmuir 25 (2009) 2575–2581, http://dx.doi.org/10.1021/la804038z J. Park, L.D. Fouché, P.T. Hammond, Multicomponent patterning of layerby-layer assembled polyelectrolyte/nanoparticle composite thin films with controlled alignment, Adv. Mater. 17 (2005) 2575–2579, http://dx.doi.org/10. 1002/adma.200501075 M. Gai, J. Frueh, A. Girard-Egrot, S. Rebaud, B. Doumeche, Q. He, Microcontact printing of PEM thin films: effect of line and surface tension, RSC Adv. (2015) (submitted for publication). F.C¸. Cebeci, D.J. Schmidt, P.T. Hammond, Multilayer transfer printing of electroactive thin film composites, ACS Appl. Mater. Interfaces 6 (2014) 20519–20523, http://dx.doi.org/10.1021/am506120e P. Zhang, J. Guan, Fabrication of multilayered microparticles by integrating layer-by-layer assembly and microcontact printing, Small 7 (2011) 2998–3004, http://dx.doi.org/10.1002/smll.201101238 X. Chen, J. Sun, J. Shen, Patterning of layer-by-layer assembled organic–inorganic hybrid films: imprinting versus lift-off, Langmuir 25 (2009) 3316–3320, http://dx.doi.org/10.1021/la804133w G. Stoychev, S. Zakharchenko, S. Turcaud, J.W.C. Dunlop, L. Ionov, Shapeprogrammed folding of stimuli-responsive polymer bilayers, ACS Nano 6 (2012) 3925–3934, http://dx.doi.org/10.1021/nn300079f G. Stoychev, S. Turcaud, J.W.C. Dunlop, L. Ionov, Hierarchical multi-step folding of polymer bilayers, Adv. Funct. Mater. (2012), http://dx.doi.org/10.1002/adfm. 201203245, n/a–n/a. M.V. Kiryukhin, S.M. Man, S.R. Gorelik, G.S. Subramanian, H.Y. Low, G.B. Sukhorukov, Fabrication and mechanical properties of microchambers made of polyelectrolyte multilayers, Soft Matter 7 (2011) 6550, http://dx.doi.org/10. 1039/c1sm05101f M.V. Kiryukhin, S.M. Man, A. Tonoyan, H.Y. Low, G.B. Sukhorukov, Adhesion of polyelectrolyte multilayers: sealing and transfer of microchamber arrays, Langmuir 28 (2012) 5678–5686, http://dx.doi.org/10.1021/la3003004 J.B. Schlenoff, S.T. Dubas, T. Farhat, Sprayed polyelectrolyte multilayers, Langmuir 16 (2000) 9968–9969. N. Kato, F. Caruso, Homogeneous, competitive fluorescence quenching immunoassay based on gold nanoparticle/polyelectrolyte coated latex particles, J. Phys. Chem. B 109 (2005) 19604–19612, http://dx.doi.org/10.1021/ jp052748f P.W. Atkins, J. de Paula, Physikalische Chemie, Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim, 2006. R. a Ghostine, M.Z. Markarian, J.B. Schlenoff, Asymmetric growth in polyelectrolyte multilayers, J. Am. Chem. Soc. 135 (2013) 7636–7646, http://dx.doi.org/ 10.1021/ja401318m R. Mueller, K. Koehler, R. Weinkamer, G. Sukhorukov, A. Fery, Melting of PDADMAC/PSS capsules investigated with AFM force spectroscopy, Macromolecules 38 (2005) 9766–9771. K. Köhler, P. Biesheuvel, R. Weinkamer, H. Möhwald, G. Sukhorukov, Salt-induced swelling-to-shrinking transition in polyelectrolyte multilayer capsules, Phys. Rev. Lett. 97 (2006) 3–6, http://dx.doi.org/10.1103/PhysRevLett. 97.188301 J. Frueh, G. Reiter, H. Moehwald, Q. He, R. Krastev, Novel controllable auxetic effect of linearly elongated supported polyelectrolyte multilayer with amorphous structure, Phys. Chem. Chem. Phys. 15 (2013) 483–488, http://dx.doi. org/10.1039/C2CP43302H J. Frueh, G. Reiter, H. Möhwald, Q. He, R. Krastev, Orientation change of polyelectrolytes in linearly elongated polyelectrolyte multilayer measured by polarized UV spectroscopy, Colloid Surf. A 415 (2012) 366–373, http://dx.doi.org/10. 1016/j.colsurfa.2012.08.070 J. Frueh, M. Gai, S. Halstead, Q. He, Chapter 2: Structure and thermodynamics of polyelectrolyte complexes, in: G.A.S. Visakh, Bayraktar, Oguz, Pico (Eds.),

278

[30] [31]

[32]

[33]

M. Gai et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 271–278 Polyelectrolyte Thermodyn. Rheol., 1st ed., Berlin, Heidelberg, Springer, 2014, p. 600. R.V. Klitzing, Internal structure of polyelectrolyte multilayer assemblies, Phys. Chem. Chem. Phys. 8 (2006) 5012–5033. K. Köhler, D.G. Shchukin, H. Möhwald, G.B. Sukhorukov, Thermal behavior of polyelectrolyte multilayer microcapsules. 1. The effect of odd and even layer number, J. Phys. Chem. B 109 (2005) 18250–18259, http://dx.doi.org/10.1021/ jp052208i J.B. Schlenoff, A.H. Rmaile, C.B. Bucur, Hydration contributions to association in polyelectrolyte multilayers and complexes: visualizing hydrophobicity, J. Am. Chem. Soc. 130 (2008) 13589–13597, http://dx.doi.org/10.1021/ja802054k H. Mjahed, J.-C. Voegel, B. Senger, A. Chassepot, A. Rameau, V. Ball, et al., Hole formation induced by ionic strength increase in exponentially growing multilayer films, Soft Matter 5 (2009) 2269, http://dx.doi.org/10.1039/b819066f

[34] Q. He, W. Song, H. Moehwald, J. Li, Hydrothermal-induced structure transformation of polyelectrolyte multilayers: from nanotubes to capsules, Langmuir 24 (2008) 5508–5513. [35] K. Köhler, H. Möhwald, G.B. Sukhorukov, Thermal behavior of polyelectrolyte multilayer microcapsules: 2. Insight into molecular mechanisms for the PDADMAC/PSS system, J. Phys. Chem. B 110 (2006) 24002–24010, http://dx.doi.org/ 10.1021/jp062907a [36] G. Decher, Q.H. Schlenoff, Joseph B. Multilayer Thin Films, 2nd ed., Wiley-VCH Verlag & Co. KGaA, Weinheim, 2012. [37] U. Micka, K. Kremer, Strongly charged flexible polyelectrolytes in poor solventsfrom stable spheres to necklace chains, Eur. Lett. 49 (2000) 189–195. [38] H.J. Limbach, C. Holm, Single-chain properties of polyelectrolytes in poor solvent, J. Phys. Chem. B 107 (2003) 8041–8055.