Microporous and Mesoporous Materials 191 (2014) 59–66

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials

journal homepage: www.elsevier .com/locate /micromeso

Fabrication of nano- to micron-sized patterns using zeolites: Itsapplication in BSA adsorption

http://dx.doi.org/10.1016/j.micromeso.2014.02.0411387-1811/� 2014 Elsevier Inc. All rights reserved.

⇑ Corresponding author at: Department of Micro and Nanotechnology, MiddleEast Technical University, 06800 Ankara, Turkey. Tel./fax: +90 312 2106459.

E-mail address: akata@metu.edu.tr (B. Akata).

Salih Kaan Kirdeciler a,b, Can Ozen b,c,d, Burcu Akata a,b,e,⇑a Department of Micro and Nanotechnology, Middle East Technical University, 06800 Ankara, Turkeyb Central Laboratory, Middle East Technical University, 06800 Ankara, Turkeyc Department of Biotechnology, Middle East Technical University, 06800 Ankara, Turkeyd Center of Excellence in Biomaterials and Tissue Engineering (BIOMATEN), Middle East Technical University, 06800 Ankara, Turkeye The Center for Solar Energy Research and Applications (GÜNAM), Middle East Technical University, 06800 Ankara, Turkey

a r t i c l e i n f o

Article history:Received 14 September 2013Received in revised form 25 December 2013Accepted 26 February 2014Available online 6 March 2014

Keywords:Zeolite patterningProtein adsorptionPhotolithographyElectron beam lithography

a b s t r a c t

Nano to micron-sized zeolite A (Z-A) and silicalite (Z-SIL) patterns were generated using the combina-tions of electron beam lithography (EBL) or photolithography (PL) with direct attachment method tobe able to generate differentiated areas on a single surface in a cheap and facile way. The possibility togenerate minimum sized zeolite patterns on top of zeolite monolayers was investigated by using EBLto understand the capability of the system for the first time. Also generation of large scale zeolite patternson top of a different zeolite monolayer was investigated by using PL allowing the generation of differen-tiated surfaces for various potential applications such as selective adsorption studies. With this combina-tion, it was shown that creating 3D zeolite architectures of different types with a perfect control in alldimensions was possible without the using any chemical linker. In order to test the potential differentbehaviors that zeolites of different properties are offering in the adsorption of biomolecules, zeolite pat-terned surfaces developed by PL were subjected to adsorption studies with bovine serum albumin (BSA).Irrespective of zeolite type, BSA always preferred the patterned regions rather than the underlying zeolitemonolayers. It can be speculated that the obtained difference in roughness values facilitates the proteinto be selectively adsorbed onto surfaces with increased roughness, i.e., the patterned regions. Moreover,we observed �2-fold fluorescence intensity difference between Z-SIL and Z-A patterns, which were sub-jected to FITC-BSA solution. Hydrophobic interactions and charge repulsion are considered as two criticalforces responsible for the observed adsorption differences.

� 2014 Elsevier Inc. All rights reserved.

1. Introduction

The design of selective coatings for advanced applications, suchas selective adsorption, sensor arrays, patterning of biomoleculesand nanoparticles, miniaturized electronics and magnetic devicesas well as development of microfluidic channels and lab-on-a-chipsystems have attracted growing attention. The spatial control ofhydrophilic/hydrophobic properties of surfaces is another parame-ter that is demanded on such surfaces [1,2]. Arrays of patterns ofnanoporous thin films are also known to be very useful as sub-strates for biological applications [3]. Several approaches are cur-rently available for preparing such surfaces of different lengthscales with varying properties that are fabricated considering the

intended usage area. The overall purpose of these studies is tochange the features on such surfaces from nanometer to microm-eter resolution with complete control in the size of the pattern fea-tures and the chemical/physical properties of the patterned andunpatterned surfaces. More challenging approach would be to cre-ate different chemical/physical properties on the same surfacewith the same control on fabrication. In general, this can be real-ized through the combination of hydrophilic/hydrophobic materi-als by lithographic means on the same surfaces [1].

Fabrication of thin films with nanopores has been of great inter-est and a difficult challenge to accomplish [3]. Attachment of dif-ferent zeolites onto surfaces in a desired way is an alternativeand promising method for production of hydrophilic/hydrophobicnanoporous regions on substrates and in that way, it is also possi-ble to benefit from the other unique properties of zeolites. Thinfilms fabricated using zeolites can bring additional advantage of in-creased surface area allowing reduction in the overall size of an

60 S.K. Kirdeciler et al. / Microporous and Mesoporous Materials 191 (2014) 59–66

array by maximizing the total number of reaction sites enabling ra-pid evaluation of different bioactivities. Furthermore, it is possibleto tailor the type of these surface groups by controlled chemical orthermal modification of these active sites leading to varying bioac-tivities [4]. Furthermore, the nanochannels of zeolites can be usedas hosts for the supramolecular organization of molecules andcomplexes, which allows the use of these materials in many differ-ent applications [5,6]. Combining different types of zeolites pos-sessing different morphologies, acidities and thus selectivity todifferent compounds on the same substrate would be another chal-lenge to achieve the above mentioned combination of physical/chemical properties on a single substrate. Such substrates wouldalso benefit from having versatile active sites with thermal andmechanical stabilities. Zeolite nanoparticles have been widelystudied in the last decade and have drawn much interest due totheir large external surface area compared with conventional zeo-lite crystals, high dispersibility in both aqueous and organic solu-tions, high thermal and hydrothermal stabilities, and tunablesurface properties such as adjustable surface charge and hydrophi-licity/hydrophobicity. The unique properties make nanozeolitespromising candidates for microfluidic surface modification andprotein immobilization.

In spite of all the benefits that zeolites and zeo-type materialscould bring into such applications, one of the main drawbacks istheir powder form upon synthesis and the need to form well or-ganized zeolite mono-multilayers on different substrates. Generalapproach for attaching zeolite microcrystals on different sub-strates is to use chemical linkers [5–8] and an excellent reviewwas reported by Yoon et al. [9]. This route usually adds an extrastep to the fabrication process leading to difficulties to adapt thedeveloped methodology to large scale areas. A relatively easierapproach would be to eliminate the chemical linker [10–13].There are also several studies performed by Yoon et al. wherezeolite micropatterned zeolite mono-multilayers were producedon glass substrates by microcontact printing [14] and photoli-thography [15]. In addition to this, the combination of electron-beam lithography (EBL) and direct attachment methodology wasshown to allow full control in zeolite film thickness generatingzeolite micropatterns with lines as small as a single zeolite nano-particle [16].

In this paper, the aim is to generate different 3D architectures ofvarying chemical/physical properties on the same substrate usingno chemical linkers. This aim was investigated through the use oftwo completely different zeolite types possessing totally differentmorphologies, pore structures, chemical and surface properties.The created zeolite nano-micro scale patterns will be a novel alter-native to selectively adsorb proteins as a function of changinghydrophilic/hydrophobic properties, the surface roughness andchemistry of zeolites in addition to the leading increased surfacearea and number of reaction sites. For that purpose, both electronbeam lithography (EBL) and photolithography techniques werecombined with direct attachment technique to produce a varietyof different line patterns of well controlled nano to micron sizes.Furthermore, FITC conjugated BSA was adsorbed on the preparedpatterns to be able to investigate the selectivity of these moleculeson differentiated areas on a single surface using two different typesof zeolites.

2. Experimental

2.1. Materials and substrates

Zeolite A nanoparticles were synthesized from a mixture havingthe following chemical composition: 11.25 SiO2:1.8 Al2O3:13.4(TMA)2O:0.6 Na2O:700 H2O Tetraethylorthosilicate (TEOS, 95%

Acros) was used as the silica source and aluminum isopropoxide(98%, Acros) was used as alumina source.

Silicalite particles were synthesized from a mixture having thefollowing composition: TPAOH:4 TEOS:350 H2O, where the silicasource was again Tetraethylorthosilicate (TEOS, 95% Acros). Theresulting solid nanoparticles were centrifuged at 13,000 rpm,washed with deionized water and dried at 80 �C. NANO950poly(-methyl methacrylate) (PMMA) C7 (average MW = 950,000,7% in chlorobenzene) was purchased from MicroChem (Newton,MA, USA) and used as electron beam (EB) resists. Methyl isobutylketone (MIBK)/Isopropanol (IPA) (1:2 (v/v)) was used as a devel-oper for EBL. Ultrapure water (>18 MO), obtained using a MESUltraPure water system, was used for all substrate cleaning steps.

P-type Si (001) wafers were subjected to thorough cleaningwithout removing the native oxide layer prior to use. After eachstep, the wafers were blow dried with dry N2 gas.

Albumin–fluorescein isothiocyanate conjugate and Fluorescein5(6)-isothiocyanate was purchased from Sigma–Aldrich.

2.2. Direct attachment of zeolites on Si wafer

The direct attachment method was used to attach zeolites on Siwafer surfaces based on the literature report [13]. Si wafers werecut into 1 cm � 1 cm pieces and placed on a piece of clean paper.About 2 mg of zeolite powder was put onto the substrates. Thenthey were pressed and rubbed onto the surface with the help ofa finger. Finally, the zeolite assembled Si wafer substrates wereheat treated at 100 �C in a conventional oven for 30 min formingthe final zeolite assembled substrates (i.e., silicalite substrates;SIL or zeolite A substrates; ZA).

2.3. Formation of zeolite patterns on monolayer of zeolites

2.3.1. Formation of zeolite patterns with EBL on monolayer of zeolites3 wt% of PMMA was obtained by diluting PMMA C7 with proper

amount of chlorobenzene. In order to form zeolite patterns of onetype on another type of zeolite assembled Si wafer substrate (i.e.,zeolite A on silicalite substrate; ZA-SIL or vice versa; SIL-ZA), 3wt% of PMMA in chlorobenzene was spun on Si wafers coated withany type of zeolite monolayer (ZA or SIL) using the above men-tioned procedure with 6000 rpm forming approximately 400 nmthick resist films (Scheme 1). After coating the resist, substrateswere pre-baked for 30 min at 160 �C. Patterns were defined by uti-lizing EBL system (Xenos XeDraw2 Pattern generator attachedCamScan CS3000 SEM). Patterned substrates were developed inMIBK/IPA solution for 60 s, rinsed in IPA, washed in flowing deion-ized water, and finally dried with N2 gas. Zeolite direct attachmentmethod was applied second time onto the PMMA coated surfaces.Prepared thin films (SIL-ZA or ZA-SIL) were put into an oven at100 �C for 30 min. Then the substrates were rinsed and ultrasoni-cated in acetone and dried using N2 gas.

2.3.2. Formation of zeolite patterns with photolithography onmonolayer of zeolites

AZ 5214 photoresist (Microchemicals) was spin coated on zeo-lite monolayers with 6000 rpm for 40 s and the resulting resistthickness was approximately 1 lm (Scheme 2). Coated zeolitemonolayers with photoresists were pre-baked at 110 �C for a min-ute. Masks were aligned on the photoresist coated wafers and uti-lized by photolithography system at GÜNAM, METU. Patternedsurfaces were developed using AZ 726 (Microchemicals) metalion free developer solution for approximately 10 s, rinsed andwashed with distilled water and dried with N2 gas flow. Directattachment methodology was applied onto the patterned surfaceswith the same manner explained in the formation of zeolitepatterns with EBL on zeolite monolayers (Section 2.3.1). Prepared

Scheme 1. Schematic illustration of the EBL procedure used in the experiments for obtaining nano-sized patterns of Z-A on Z-SIL modified substrates and vice versa.

Scheme 2. Schematic illustration of the PL procedure used in the experiments for obtaining micron-sized patterns of Z-A on Z-SIL and vice versa (after direct attachmentprocedure shown in Scheme 1).

S.K. Kirdeciler et al. / Microporous and Mesoporous Materials 191 (2014) 59–66 61

thin films (SIL-ZA or ZA-SIL) were put into an oven at 100 �C for30 min. Then the substrates were rinsed and ultrasonicated in ace-tone and dried using N2 gas.

2.4. Fluorescent molecule adsorption on zeolite patterns

Fluorescein isothiocyanate labelled bovine serum albumin(FITC-BSA) were simply adsorbed onto the zeolite patterns byimmersing zeolite samples in 0.5 mg/ml FITC-BSA suspensionsfor an hour under stirring conditions. The samples were then takenout from the suspension and were washed six times with phos-phate buffered saline solution (PBS) and rinsed with Ultrapurewater (>18 MO). Fluorescence intensity was measured using a con-focal laser scanning microscope (LSM510, Carl Zeiss Micro ImagingGmbH, Jena, Germany) with Plan-Neofluar 20� objective (NA 0.5),488 nm argon laser excitation and LP505 emission filter.

3. Results and discussion

3.1. Obtaining zeolite monolayers on Si wafer

The most efficient method for zeolite assembly on silicon waferwithout using any chemical linker was found to be direct attach-ment methodology [16]. In the current study, zeolite A (Z-A; Si/Al ratio of �1) and silicalite (Z-SIL; Si/Al ratio of 1) monolayerswere formed successfully using this methodology without usingany chemical linker as before. Z-A and Z-SIL have been widelyinvestigated due to their changing Si/Al ratio and thus their

hydrophobic characters [17]. Accordingly, direct attachment tech-nique was used to obtain facile zeolite monolayers of Z-SIL and Z-Aon Si substrates, where the whole fabrication can be feasible inonly several minutes. The obtained zeolite modified substrates ofZ-SIL and Z-A are shown in Fig. 1.

In order to check different hydrophobic characters of the ob-tained zeolite modified substrates, contact angle measurementswere performed on the surfaces shown in Fig. 1a and b. Accord-ingly, Z-SIL and Z-A modified surfaces have contact angles of96.6 ± 2.2 and 71.3 ± 2 respectively. It is well known that Si/Al ratiois used to denote hydrophobicity, with higher ratios indicating ahigher degree of hydrophobicity [18]. It was confirmed that theZ-SIL monolayer with higer Si/Al ratio was more hydrophobic withrespect to its Z-A monolayer counterpart for the first time. Theseresults suggest that zeolite modification of substrates can be usedas a facile approach to obtain different chemical affinities on anygiven substrate without using any chemical linker. Futhermore,the proposed procedure of zeolite modification enables a reproduc-ible preparation on centimeter scale which can be of importancefor industries where preparation on large scale is especiallyimportant.

3.2. Fabrication of different types of zeolites on the same substrate byEBL

We had previously shown that it was possible to fabricate Z-Apatterns on Si substrate with the features of the patterned stripesof as small as the zeolite size (200–250 nm) by combining direct

Fig. 1. SEM images of the monolayers of Z-SIL (a) and Z-A (b) prepared on Si substrates using direct attachment methodology.

Fig. 2. SEM images of monolayers of single (a) and double Z-SIL (b) line patterns on Z-A monolayer on Si substrate prepared using direct attachment methodology and EBL.

62 S.K. Kirdeciler et al. / Microporous and Mesoporous Materials 191 (2014) 59–66

attachment method and EBL [16]. It is known that the nature ofbonding between zeolite crystals and substrates are of ionic andhydrogen bonding for assembling monolayers by direct attach-ment technique [13]. This time, we built that architecture on an-other zeolite layer. SEM micrographs of single and double Z-SILline patterns on Z-A monolayer substrates can be seen in Fig. 2.

According to Fig. 2, substrates that retain two different zeolitetypes where each of them possessing its own unique properties,such as Si/Al ratio and morphologies, was accomplished with suchcontrolled features for the first time. This approach offers an alter-native to one of the very important issues of nanofabrication,which is precise control of chemical composition of the creatednano-patterns. This is even more feasible with the many alterna-tives to choose different zeolite types and the ability to modifythe surface –OH groups of zeolites according to the end use of suchfabricated surfaces. Such a combination was achieved previouslyby Sun et al. by first growing one type of zeolite (ZSM-5) on Si wa-fer after which the second type (Z-SIL) was grown on the fabricatedand etched ZSM-5 layer with 5–70 lm features [17]. However, thisprocedure of growing zeolite films on substrates involves manysteps and necessitates the use of various different chemicals, etch-ing, and elevated temperatures (i.e., above 100 �C). Such proce-dures limit the choice of substrates and their actual integrationinto device parts which may involve polymeric connections. Fur-thermore, the ability to tailor the zeolite surface precisely, that isonly the patterned features or vice versa, is not possible when zeo-lite is grown on substrates. In the current study, such small andwell controlled distribution in nano-features of zeolite patterns

on a monolayer of another distinct zeolite type was achieved forthe first time with the additional benefit of easy, custom-built pro-cedure that involves no chemicals and is applicable to varying sub-strates. Various examples of zeolite patterns that are of differenttypes than the ones used to make monolayers (i.e., Z-A patternson Z-SIL monolayers: Fig. 3a–d; Z-SIL patterns on Z-A monolayers:Fig. 3e–h) can be seen in Fig. 3.

Although such practical zeolite modified surfaces wereachieved in a very controlled manner successfully, EBL techniquestill suffers due to several reasons. First of all, this technique iscomplex and costly. Secondly, a reproducible fabrication in largerscales is very difficult and time consuming. However, large areafabrication, applicability to various substrates, as well as fabricat-ing patterns with enlarged features (i.e., larger than 2–10 lm)can be of great interest, especially for the development of newapplications in biomaterial science and biotechnology. Therefore,further improvement of this technique was examined by usingphotolithography to actually test the potential use of this approachthat combines two different layers of zeolite on the same surface,where one layer is of different zeolite type and patterned into var-ious forms for the first time.

3.3. Fabrication of different types of zeolites on the same substrate byphotolithography

The combination of direct attachment method of zeolites andphotolithography (PL) was believed to offer an alternative to oneof the main challenges in nanolithography, which is to precisely

Fig. 3. SEM images of Z-A patterns on Z-SIL monolayers (a–d); Z-SIL patterns on Z-A monolayers (e–h) prepared using direct attachment methodology and EBL.

Fig. 4. SEM images of Z-A patterns on Z-SIL monolayers (a–c); Z-SIL patterns on Z-A monolayers (d–f) prepared using direct attachment methodology and PL.

S.K. Kirdeciler et al. / Microporous and Mesoporous Materials 191 (2014) 59–66 63

control the chemical composition, size, shape and distribution ofcreated nano-features and to be a low-cost, high-throughput pro-duction method [19].

Accordingly, direct attachment method was combined with PL ina similar fashion with the EBL approach in order to fabricate twodifferent types of zeolites, thus producing different chemical andstructural features on the same surface. Scheme 2 shows the methodwe have developed by PL to generate zeolite patterns of one type onanother type of zeolite monolayer (i.e., Z-A patterns on Z-SIL mono-layer, Fig. 4a–c; and Z-SIL patterns on Z-A monolayer, Fig. 4d–f). Theonly variation in the process for PL with respect to EBL is that thepatterns were generated using a mask, while the pattern to be rep-licated and exposure of the whole surface to UV light since the resist

used (AZ5214) is sensitive to this light. Representative SEM imagesof the obtained surfaces obtained after combining direct attachmentand PL are shown in Fig. 4.

As shown in Fig. 4a–f, the fabrication route that was followedgenerated versatile, well-defined Z-A and Z-SIL patterns on theunderlying zeolite monolayers. The pattern widths were of 5 lm,whereas the groove sizes were significantly larger than the onesgenerated using EBL, which were around 50 lm. These featuresizes were chosen to allow us to test these substrates in BSAattachment studies, since the total patterned area generated withEBL is very small (50 lm � 50 lm). Although 5 lm grooves couldbe attained using PL, the process of direct attachment gets harderto achieve as the groove size is reduced. This is due to the high

64 S.K. Kirdeciler et al. / Microporous and Mesoporous Materials 191 (2014) 59–66

possibility to damage and clean the resist during the direct attach-ment process, which is simply finger pressing of zeolite powderonto the substrate. This was not such a big problem during EBL, be-cause the resist used during EBL was seen to possess enhancedattachment and stability with respect to PL resist. Accordingly,the stability of the whole process can be a function of the resistused. Furthermore, total patterned area was also significantly lar-ger and easy to process with respect to EBL technique, where onewhole Si wafer could be easily fabricated in only a couple of sec-onds. Thus, PL allowed us to actually immobile biological mole-cules on the surfaces generated and monitor these molecules(BSA in this particular study, vide infra) by fluorescence micro-scope (Scheme 2).

Another distinct feature of the zeolite patterns generated usingPL was the higher thickness of these patterns with respect to EBL.The thickness of the patterns obtained using PL were of �1 lm,which corresponds to around 4–5 zeolite monolayers (Fig. 4c andf), whereas they were of a single monolayer using EBL (Fig. 3). Itis almost impossible to obtain patterns of a single monolayer thick-ness, since the photoresist used for PL process is a lot more viscousthan the EBL resist. Thus, although the same procedure of directattachment was applied for PL onto the resist coated and devel-oped substrate, it was seen that the dominating factor for the pat-tern width was the resist thickness. This was actuallydemonstrated in our previous study that the resist thickness wasused to control the number of zeolite layers where a maximumof two monolayers were achieved on the Si substrate. In this par-ticular study, it was further shown that the PL resist used wasaround �1 lm and thus it actually resulted into 4–5 zeolite layerthick patterns, each zeolite being 250 nm wide, in correlation withour previous study [16].

In comparison with PL technique, EBL technique has a betterresolution in z-direction due to the resist thickness, and this

Table 1SEM images with the AFM images taken from the areas as marked by red boxes and their

thinner layer achieved by EBL leads us to produce single zeolitethick patterns. Furthermore, the zeolite patterns prepared withPL methodology, has 4–5 zeolite layers on patterns, and this multi-ple layers cause a rough surface with respect to the monolayer ly-ing down on the bottom. To be able to investigate the effect ofmultiple layers on surface roughness and topography, atomic forcemicroscopy experiments were carried out. The representativeimages and surface roughness data of the samples were given inTable 1. As expected, zeolite monolayers have a very smooth sur-face with respect to the patterned regions that consist of multiplelayers. More interestingly, even if the zeolite type changed (Z-A/Z-SIL), no significant difference in surface roughness was observed.

3.4. Protein adsorption on zeolite patterns

Factors influencing the interactions between the surface andproteins have been of great interest in understanding the adsorp-tion mechanism of guest molecules on such surfaces [19,20]. In or-der to test the potential different behaviors that zeolites ofdifferent properties are offering in the adsorption of biomolecules,zeolite patterned surfaces developed in the current study weresubjected to adsorption studies with bovine serum albumin(BSA). Most of the literature work in this area used different typesof zeolites and performed adsorption studies in solution phase,where zeolite nanocrystals were not necessarily attached ontoany substrate [20,21]. However, the interactions between proteinsand solid surfaces can also be effected by the heterogeneity of thesurface and molecular properties of the adsorbed molecules [22].In that sense, it is believed that the developed surfaces offer a greatdeal of potential to open up the gate for investigating suchinteractions.

In literature, the adsorption of proteins onto zeolites is knownto be selective as a result of several factors that are regulated

representative surface roughness values (Ra).

S.K. Kirdeciler et al. / Microporous and Mesoporous Materials 191 (2014) 59–66 65

mainly according to the isoelectric point (pI) of the proteins. Thesecan be Coulombic attraction below pI; hydrophobic interactionsand mesopore structure at pI; and hydrophobic interaction abovepI [19]. These factors are reported based on the studies performedin solution. However, different factors might also influence theideal adsorption characteristics and further complicate the inter-pretation of results, such as the asymmetric charge distributionon the BSA surface [23], and the heterogeneity and/or the surfaceroughness of the adsorbed surfaces [24]. Although it can be a diffi-cult task to pin-point the exact factor influencing the interactionbetween the protein and the surface, a more difficult task wouldbe to design such a surface that retains various parameters onthe same substrate which serves for a particular demand. In orderto test whether the zeolite modified and patterned substrates showany type of selectivity to BSA adsorption, the ones that were fabri-cated using PL were chosen in the current study. Patterns gener-ated using EBL was not applicable for this study, since theprepared patterns are too small to observe (50 lm in total pat-terned area) the differences between different adsorption charac-teristics of the patterned and monolayer zeolite regions (Fig. 3)by fluorescence microscopy technique.

To test whether surface modification with zeolites resulted inany selectivity in protein adsorption, the Si substrates patternedwith Z-SIL on Z-A monolayer and Z-A with Z-SIL monolayer sam-ples (Fig. 4 and Table 1) were subjected to BSA adsorption studies.Protein adsorption capacity of SIL and ZA zeolites were comparedusing FITC-conjugated Bovine Serum Albumin (BSA). Following a0.5 mg/ml BSA solution treatment, patterned zeolite surfaces werewashed with the PBS buffer. Fluorescence signal intensity wasmeasured with a confocal laser scanning microscope.

As shown in Fig. 5a and b, BSA prefers the 5 lm width patternedsurfaces rather than the underlying monolayer zeolite regions of50 lm. Irrespective of zeolite type, BSA was observed to alwaysprefer this behavior. Accordingly, distinct chemical properties of

Fig. 5. Confocal fluorescence microscopy images and 3D intensity plots (X and Y in lm,Surface charge distribution of BSA and critical molecular forces between the protein an

different zeolite types did not seem to affect BSA adsorption be-tween the monolayer and patterned regions. It was shown thatthe major difference between the monolayers and patterned re-gions was their difference in their roughness values (Table 1). Pat-terned regions were shown to have �5 times larger roughness thantheir monolayer counterparts. Accordingly, it can be speculatedthat the obtained difference in roughness values facilitates the pro-tein to be selectively adsorbed onto surfaces with increased rough-ness. It was actually shown that roughness on the nanometer scalehas significant impact on protein adsorption and therefore rough-ness is an important parameter in biomaterials design [24,25]. Inthe current study, the thickness of the photoresist allowed the con-trol of zeolite thickness of the patterned regions, which was around4–5 times thicker (i.e., 4–5 times increased number of zeolite lay-ers corresponding to the �1.5 lm thick photoresist) than the mon-olayers (�250 nm). Therefore, creating zeolite patterns can behypothesized to offer an alternative method for control in rough-ness values offering an additional impact in protein adsorption.

Moreover, we observed �2-fold fluorescence intensity differ-ence between Z-SIL and Z-A patterns, which indicates strongerBSA adsorption on the Z-SIL zeolite patterns (Fig. 5). BSA has an iso-electric point (pI) of 4.7 and carries a net negative charge at pH 7.2.It is also well known that zeolites have a net negative charge,which can be stronger upon the partial substitution of Si4+ ionsin zeolites with Al3+. Accordingly, Z-A zeolite surface is negativelycharged due to their higher Al content. Therefore, weaker proteinadsorption observed on the Z-A pattern may be originating fromcharge repulsion. Furthermore, Si/Al ratio is used to denote thehydrophobicity of zeolites with higher ratios indicating a higherdegree of hydrophobicity [19]. Thus, compared to Z-A, Z-SIL has amore hydrophobic nature. BSA has also hydrophobic patchesdistributed over its surface and higher protein adsorption on theZ-SIL pattern can be explained by favorable hydrophobic interac-tions. Overall interaction energy between a protein molecule and

intensity as RFU) of FITC conjugated BSA adsorbed on Z-SIL (a) and Z-A (b) patterns.d zeolite surfaces are shown in (c).

66 S.K. Kirdeciler et al. / Microporous and Mesoporous Materials 191 (2014) 59–66

a surface is a combination of various favorable and unfavorableforces including electrostatic interactions, hydrogen bonding andVan der Waals forces. In our case, favorable hydrophobic interac-tions and unfavorable charge repulsion seem to be the criticalforces in determining the overall strength of BSA-(Z-SIL) andBSA-(Z-A) interactions respectively (Fig. 5c).

4. Conclusions

In the current study, as small as a single (�250 nm) to severalzeolite thick patterns on a monolayer of another distinct zeolitetype, i.e., Z-SIL patterns on Z-A monolayer and Z-A patterns onZ-SIL monolayer, were fabricated by EBL and PL techniques respec-tively for the first time. The fabrication route that consists ofcombining direct attachment with lithographic techniques offersan additional benefit of easy, custom-built procedure that involvesno chemicals and can be easily applicable to varying substrates. Totest whether surface modification with zeolites resulted in anyselectivity in protein adsorption, Si substrates patterned withdistinct zeolite features were subjected to FITC-conjugated BovineSerum Albumin (BSA). It was seen that, irrespective of zeolite type,BSA prefers the 5 lm width patterned surfaces rather than theunderlying monolayer zeolite regions of 50 lm. Accordingly, rough-ness of the surfaces seem to govern protein adsorption and distinctchemical properties of different zeolite types did not seem to have anaffect between the monolayer and patterned regions. However,�2-fold fluorescence intensity difference was observed betweenZ-SIL and Z-A patterns indicating stronger BSA adsorption on theZ-SIL zeolite patterns that are known to possess a more hydrophobicnature than Z-A. In summary, protein adsorption studies indicatedfavorable hydrophobic interactions and unfavorable charge repul-sion that determines the protein-zeolite interactions on solid sur-faces that are also governed by the surface roughness properties.

Acknowledgments

The support provided by METU-Central Laboratory and METU-GUNAM is greatly acknowledged.

References

[1] T. Kobayashi, K. Shimizu, Y. Kaizuma, S. Konishi, Lab Chip 11 (2011) 639.[2] L. Yan, K. Wang, J. Wu, L. Ye, Eng. Aspects 296 (2007) 123.[3] H.C. Kim, G. Wallraff, C.R. Kreller, S. Angelos, V.Y. Lee, W. Volksen, R.D. Miller,

Nano Lett. 4 (2004) 1169.[4] E. Soy, V. Arkhypova, O. Soldatkin, M. Shelyakina, S. Dzyadevych, J.

Warzywoda, A. Sacco, B. Akata, Mater. Sci. Eng: C 32 (2012) 1835.[5] N.S. Kehr, K. Riehemann, J. El-Gindi, A. Schafer, H. Fuchs, H.J. Galla, L. De Cola,

Adv. Funct. Mater. 20 (2010) 2248.[6] G. Calzaferri, Langmuir 28 (2012) 6216.[7] A.Z. Ruiz, H. Li, G. Calzaferri, Angew. Chem. Int. Ed. 45 (2006) 5282.[8] S. Hashimoto, K. Samata, T. Shoji, N. Taira, T. Tomita, S. Matsuo, Microporous

Mesoporous Mater. 117 (2009) 220.[9] K.B. Yoon, Acc. Chem. Res. 40 (2007) 29.

[10] P. Laine, R. Seifert, R. Giovanoli, G. Calzaferri, New J. Chem. 21 (1997) 453.[11] C. Leiggener, G. Calzaferri, ChemPhysChem 5 (2004) 1593.[12] S. Mintova, T. Bein, Adv. Mater. 13 (2001) 1880.[13] J.S. Lee, J.H. Kim, Y.J. Lee, N.C. Jeong, K.B. Yoon, Angew. Chem. 119 (2007) 3147.[14] K. Ha, Y.J. Lee, D.Y. Jung, J.H. Lee, K.B. Yoon, Adv. Mater. 12 (2000) 1614.[15] K. Ha, Y.J. Lee, Y.S. Chun, Y.S. Park, G.S. Lee, K.B. Yoon, Adv. Mater. 13 (2001)

594.[16] S. Ozturk, B. Akata, Microporous Mesoporous Mater. 126 (2009) 228.[17] W. Sun, K.F. Lam, L.W. Wong, K.L. Yeung, Chem. Commun. 39 (2005) 4911.[18] M. Matsui, Y. Kiyozumi, T. Yamamoto, Y. Mizushina, F. Mizukami, K. Sakaguchi,

Chemistry 7 (2001) 1555.[19] F.A. Denis, P. Hanarp, D.S. Sutherland, Y.F. Dufrene, Langmuir 20 (2004) 9335.[20] A. Tavolaro, P. Tavolaro, E. Drioli, Colloids Surf. B 55 (2007) 67.[21] E. Soy, V. Pyeshkova, V. Arkhypova, B. Khadro, N. Jaffrezic-Renault, A. Sacco,

S.V. Dzyadevych, B. Akata, Sensor Electron. Microsyst. Technol. 1 (2010) 28.[22] W. Norde, Colloids Surf. B 61 (2008) 1.[23] B. Jachimska, A. Pajor, Bioelectrochemisty 87 (2012) 138.[24] K. Rechendorff, M.B. Hovgaard, M. Foss, V.P. Zhdanov, F. Besenbacher,

Langmuir 26 (2006) 10885.[25] D.D. Deligianni, N.D. Katsala, P.G. Koutsoukos, Y.F. Missirlis, Biomaterials 22

(2000) 87.