10322 DOI: 10.1021/la901161z Langmuir 2009, 25(17), 10322–10328Published on Web 06/18/2009

pubs.acs.org/Langmuir

© 2009 American Chemical Society

Antimicrobial Behavior of Polyelectrolyte-Surfactant Thin FilmAssemblies

Charlene M. Dvoracek,† Galina Sukhonosova,† Michael J. Benedik,‡ and Jaime C. Grunlan*,†

†Department of Mechanical Engineering, Texas A&M University, College Station, Texas 77843, and‡Department of Biology, Texas A&M University, College Station, Texas 77843

Received April 2, 2009. Revised Manuscript Received May 18, 2009

Layer-by-layer (LbL) assembly, a technique that alternately deposites cationic and anionic materials, has proven tobe a powerful technique for assembling thin films with a variety of properties and applications. The present workincorporates the antimicrobial agent cetyltrimethylammonium bromide (CTAB) in the cationic layer and uses poly(acrylic acid) (PAA) as the anionic layer.When the films are exposed to a humid environment, these agents diffuse out ofthe film, inhibiting bacterial growth in neighboring regions. Film growth, microstructure, and antimicrobial efficacy arestudied here, with 10-bilayer films yielding thicknesses on the order of 2 μm. Various factors are shown to influence theantimicrobial efficacy including time, temperature, secondary ingredients, and number of bilayers. As more layers aredeposited, antimicrobial efficacy is increased because more CTAB is able to diffuse throughout the film, and higheramounts of antimicrobials are released. Additionally, inclusion of the cationic poly(diallyldimethylammonium chloride)(PDDA) in the cationic layer in conjunction with CTAB increases film uniformity, and as a result, antimicrobialeffectiveness is enhanced. These thin films provide the ability to render a surface antimicrobial and may be useful forbandages or sterilization of disposable objects (e.g., surgical marker).

Introduction

Langmuir’s discovery that surfaces will adsorb only a singlelayer of ions initiated the field of thin films with thicknesses in thenanometer to micrometer range.1 Iler built upon this concept bycreating multilayer films with positively and negatively chargedparticles.2 In the early nineties,Decher further developed this ideaby creating the formal layer-by-layer (LbL) assembly process.3-5

In this process, a substrate is alternately dipped into aqueoussolutions containing charged ingredients, as shown in Figure 1,building a film through electrostatic attractions. Hydrogen bond-ing and other types of van der Waals attractions can also be usedto build LbL assemblies,6-8 but electrostatic-based depositionremains the predominant form.2,9-12 Each positive and negativepair deposited is known as a bilayer (BL), with each bilayertypically 1-100 nm thick.9,13 These films are highly tailorable byaltering pH,11,14 ionic strength,10,14 chemistry,15 and molecularweight.12,16 Additionally, film properties can be tailored by

adding small concentrations of additives to the deposition solu-tion. These additives include clay,17-19 viruses,20 colloidal parti-cles,21,22 or antimicrobial agents.23,24

The layer-by-layer technique has been used to make thin filmsfor antireflection,25 gas barrier,18 battery electrolytes,26 andmicrocapsule drug delivery.27,28 The incorporation of smallmolecules or nanoparticles, added to either the cationic or anionicmixtures, can impart different characteristics or properties to thefilm23 or to the surroundings, including diffusion of differenttypes of biomolecules from the films.29,30 For example, theaddition of antiseptic agents could be useful in applications suchas food packaging31-34 and wound dressing.35 Silver particles areknown to kill a broad array of infectious bacteria, making them

*To whom correspondence should be addressed. Tel: +1 979 845 3027.Fax: +1 979 862 3989. E-mail: jgrunlan@tamu.edu.(1) Langmuir, I. Method of substance detection. General Electric Co.;

U.S. Patent 2232539, Feb 1941.(2) Iler, R. K. J. Colloid Interface Sci. 1966, 21, 569–594.(3) Decher, G.; Hong, J. D. Int. J. Phys. Chem. 1991, 95, 1430–1434.(4) Decher, G.; Hong, J. D.; Schmitt, J. Thin Solid Films 1992, 210, 831–835.(5) Hammond, P. T. Adv. Mater. 2004, 16, 1271–1293.(6) DeLongchamp, D. M.; Hammond, P. T. Langmuir 2004, 20, 5403–5411.(7) Fu, Y.; Bai, S. L.; Cui, S. X.; Qiu, D. L.; Wang, Z. Q.; Zhang, X.

Macromolecules 2002, 35, 9451–9458.(8) Kotov, N. A. Nanostruct. Mater. 1999, 12, 789–796.(9) Decher, G.; Schlenoff, J. B. Multilayer Thin Films: Sequential Assembly of

Nanocomposite Materials; Wiley-VCH: Weinheim, Germany, 2003.(10) McAloney, R. A.; Sinyor, M.; Dudnik, V.; Goh, M. C. Langmuir 2001, 17,

6655–6663.(11) Shiratori, S. S.; Rubner, M. F. Macromolecules 2000, 33, 4213–4219.(12) Sui, Z. J.; Salloum, D.; Schlenoff, J. B. Langmuir 2003, 19, 2491–2495.(13) Jan, C. J.; Walton, M. D.; McConnell, E. P.; Jang, W. S.; Kim, Y. S.;

Grunlan, J. C. Carbon 2006, 44, 1974–1981.(14) Schoning,M. J.; H.A.,M.; Poghossian, A. J. Solid State Electrochem. 2009,

13, 115–122.(15) Mermut, O.; Barrett, C. J. J. Phys. Chem. B 2003, 107, 2525–2530.(16) Zhang, H. Y.; Wang, D.; Wang, Z. Q.; Zhang, X. Eur. Polym. J. 2007, 43,

2784–2791.

(17) Eckle, M.; Decher, G. Nano Lett. 2001, 1, 45–49.(18) Jang, W. S.; Rawson, I.; Grunlan, J. C. Thin Solid Films 2008, 516, 4819–

4825.(19) Kotov, N. Mater. Perform. 2008, 47, 20–21.(20) Yoo, P. J.; Nam, K. T.; Qi, J. F.; Lee, S. K.; Park, J.; Belcher, A. M.;

Hammond, P. T. Nat. Mater. 2006, 5, 234–240.(21) Dawidczyk, T. J.; Walton, M. D.; Jang, W.-S.; Grunlan, J. C. Langmuir

2008, 24, 8314–8318.(22) Sukhorukov, G. B.; Donath, E.; Lichtenfeld, H.; Knippel, E.; Knippel, M.;

Budde, A.; Mohwald, H. Colloids Surf., A 1998, 137, 253–266.(23) Grunlan, J. C.; Choi, J. K.; Lin, A. Biomacromolecules 2005, 6, 1149–

1153.(24) Podsiadlo, P.; Paternel, S.; Rouillard, J. M.; Zhang, Z. F.; Lee, J.; Lee, J.

W.; Gulari, L.; Kotov, N. A. Langmuir 2005, 21, 11915–11921.(25) Fujita, S.; Shiratori, S. Jpn. J. Appl. Phys., Part 1 2004, 43, 2346–2351.(26) Lowman, G. M.; Tokuhisa, H.; Lutkenhaus, J. L.; Hammond, P. T.

Langmuir 2004, 20, 9791–9795.(27) Antipov, A. A.; Sukhorukov, G. B.; Donath, E.; Mohwald, H. J. Phys.

Chem. B 2001, 105, 2281–2284.(28) Wood, K. C.; Little, S. R.; Langer, R.; Hammond, P. T.Angew. Chem., Int.

Ed. 2005, 44, 6704–6708.(29) Pilbat, A. M.; Szegletes, Z.; Kota, Z.; Ball, V.; Schaaf, P.; Voegel, J. C.;

Szalontai, B. Langmuir 2007, 23, 8236–8242.(30) Richert, L.; Lavalle, P.; Payan, E.; Shu, X. Z.; Prestwich, G.D.; Stoltz, J. F.;

Schaaf, P.; Voegel, J. C.; Picart, C. Langmuir 2004, 20, 448–458.(31) Becerril, R.; Gomez-Lus, R.; Goni, P.; Lopez, P.; Nerin, C. Anal. Bioanal.

Chem. 2007, 388, 1003–1011.(32) Han, J. H. Food Technol. 2000, 54, 56–65.(33) Quintavalla, S.; Vicini, L. Meat Sci. 2002, 62, 373–380.(34) Tripathi, S.; Mehrotra, G. K.; Dutta, P. K. E-Polymers; No. 93; 2008.(35) Sant, S. B.; Gill, K. S.; Burrell, R. E. Philod. Mag. Lett. 2000, 80, 249.

DOI: 10.1021/la901161z 10323Langmuir 2009, 25(17), 10322–10328

Dvoracek et al. Article

the most widely used antimicrobial agent.36-40 Other widelyused antiseptics include iodine,41 quaternary ammonium com-pounds,42,43 essential oils,44-47 antibiotics,48,49 and cetyltri-methylammonium bromide (CTAB),23,50,51 which is the focusof this work. The antimicrobial agents can be incorporated intheir ionic form in layer-by-layer assemblies because theircounterions (Br- in the case of CTAB) are displaced duringdeposition. Evidence suggests that antimicrobial action occursprimarily in this charged state,52,53 eliminating the need for anactivation step. This could provide greater effectiveness atlower concentration than in other systems where antisepticsare incorporated as uncharged solids or salts.23

In the present study, CTAB is incorporated into the cationiclayers of a film, using LbL assembly. CTAB has been shownpreviously to demonstrate greater antimicrobial efficacy in LbLfilms than silver;23 this partnered with its potential for growthmade it suitable for further investigation. Growth trends of thevarious recipes for films were studied, and the final film cross-sections and surfaces were examined. Next, the effects of thenumber of bilayers, antimicrobial concentration, incubationtemperature, chemistry, and time delay after deposition wereexplored with respect to antimicrobial effectiveness. Efficacy ofthese films, measured with a Kirby-Bauer-like test, is strongestwith more bilayers, lower testing temperatures, and when usingPDDA in combination with CTAB in the cationic layers.

Materials and Methods

The anionic deposition solution consisted of 0.2 wt % poly(acrylic acid) (PAA) (Aldrich, St. Louis, MO) with a molecularweight (Mw) of 100,000-200,000 g/mol in deionized water(18.2 MΩ). Cationic solutions contained 0.2 wt % poly(diallyldi-methylammonium chloride) (PDDA) (Aldrich, St. Louis, MO)unless otherwise noted. The antimicrobial agent, cetyltrimethy-lammonium bromide (Aldrich, St. Louis, MO), was added to thecationic solution at various molarities to determine its maximumeffectiveness. All solutions were used at their natural pH; typicalvalues are 2.7 for 0.2 wt%PAA, 4.8 for 0.2 wt%PDDA, and 5.1for 0.2 wt % PDDA with 5 mM CTAB. Substrates used indifferent applications were 175 μm poly(ethylene terephthalate)(PET) (trade name ST505 by DuPont Teijin, Tekra Corp., NewBerlin, WI), polystyrene (PS) (Goodfellow, Oakdale, PA), andsilicon wafers polished on one side (University Wafer, SouthBoston,MA). Bacterial growthmedia were fromDifco LB Brothsolidified with 1.5% bacteriological agar (United States Biologi-cals, Swampscott, MA). Escherichia coli (E. coli) K-12, lab strainMB 458, (Fh galK16 galE15 relA1 rpsL150 spoT1 mcrB1) andStaphylococcus aureus (S. aureus) wild type strain, lab strain MB1594, were the bacteria used in testing.

Film Deposition. In all cases, the substrate was negativelycharged, either by using a substrate with an inherent negativecharge or corona treatment (BD-20C Corona Treater, Electro-Technic Products Inc., Chicago, IL) of an uncharged polymersubstrate, such as PET. Corona treatment oxidizes the polymeric

Figure 1. Schematic of the LbL process involving alternate dipping in cationic and anionic solutions, with rinsing and drying between eachdeposition. A schematic of the resulting thin filmmade fromCTABmolecules and poly(acrylic acid), as the positively and negatively chargedmolecules, is also shown.

(36) Chambers, H. F. In Goodman and Gilman’s The Pharmacological Basis ofTheraputics, 10th ed; Hardman, J. G., Gilman, A. G., Ed.; McGraw-Hill: NewYork, 2001.(37) Balogh, L.; Swanson, D. R.; Tomalia, D. A.; Hagnauer, G. L.; McManus,

A. T. Nano Lett. 2001, 1, 18–21.(38) Kumar, R.; Munstedt, H. Biomaterials 2005, 26, 2081–2088.(39) Redmond, S. M.; Rand, S. C.; Tang, H. X.; Martin, D. C.; Balogh, P.;

Balogh, L. Abstracts of Papers of the American Chemical Society; AmericanChemical Society: Washington, DC, 2000; Vol. 220, pp U285-U286.(40) Sondi, I.; Salopek-Sondi, B. J. Colloid Interface Sci. 2004, 275, 177–182.(41) Touitou, E.; Deutsch, J.; Matar, S. Int. J. Pharm. 1994, 103, 199–202.(42) Gottenbos, B.; van der Mei, H. C.; Klatter, F.; Nieuwenhuis, P.; Busscher,

H. J. Biomaterials 2002, 23, 1417–1423.(43) Tiller, J. C.; Liao, C. J.; Lewis, K.; Klibanov, A. M. Proc. Natl. Acad. Sci.

U.S.A. 2001, 98, 5981–5985.(44) Lopez, P.; Sanchez, C.; Batlle, R.; Nerin, C. J. Agric. Food Chem. 2005, 53,

6939–6946.(45) Lopez, P.; Sanchez, C.; Batlle, R.; Nerin, C. J. Agric. Food Chem. 2007, 55,

8814–8824.(46) Rodriguez, A.; Batlle, R.; Nerin, C. Prog. Org. Coat. 2007, 60, 33–38.(47) Rodriguez, A.; Nerin, C.; Batlle, R. J. Agric. Food Chem. 2008, 56, 6364–

6369.(48) Donelli, G.; Francolini, I.; Piozzi, A.; Di Rosa, R.; Marconi, W. J.

Chemother. 2002, 14, 501–507.(49) Gu, H. W.; Ho, P. L.; Tong, E.; Wang, L.; Xu, B.Nano Lett. 2003, 3, 1261–

1263.(50) Evans, D. J.; Allison, D. G.; Brown, M. R. W.; Gilbert, P. J. Antimicrob.

Chemother. 1990, 26, 473–478.(51) Brown, M. R. W.; Collier, P. J.; Gilbert, P. Antimicrob. Agents Chemother.

1990, 34, 1623–1628.(52) Dibrov, P.; Dzioba, J.; Gosink, K. K.; Hase, C. C. Antimicrob. Agents

Chemother. 2002, 46, 2668–2670.(53) Simonetti, N.; Simonetti, G.; Bougnol, F.; Scalzo, M. Appl. Environ.

Microbiol. 1992, 58, 3834–3836.

10324 DOI: 10.1021/la901161z Langmuir 2009, 25(17), 10322–10328

Article Dvoracek et al.

surface, providing better cationic adhesion and higher repeat-ability.54,55 Prior to corona treatment, the PET and PS filmswere rinsed with methanol and deionized water, and thendried with filtered air, while glass and silicon substrates wererinsed with acetone instead of methanol. The cleaned sub-strates were then dipped alternately in positive and negativesolutions to build the film. The initial dip in each solution wasfive minutes, with subsequent dips of one minute each. Be-tween each layer, the films were rinsed with deionized waterand blown dry with air. Specimens were stored in a desiccatorprior to testing.

Film Growth. A Maxtek Research Quartz Crystal Microba-lance (RQCM) from Infinicon (East Syracuse, NY) with afrequency range of 3.8-6 MHz was used in conjunction with5 MHz quartz crystals. The crystal, in its holder, was dippedalternately in the positive (PDDA + CTAB or CTAB only) andnegative (0.2 wt%PAA) solutions, with frequency (which is latercorrelated to mass) measured at every layer. A Dektak 3 StylusProfilometer (Neutronix-Quintel, Morgan Hill, CA) was alsoused to more directly measure thickness. Films evaluated usingprofilometry were deposited onto glass slides. This method givesan absolute measurement of thickness, but it is not very accuratefor film thicknesses below 1 μm. The profilometry readings weretaken upon the completion of each film, while the QCMmeasure-ments were used to monitor the film’s growth.

Film Characterization. Film surfaces were imaged with aNanosurf EasyScan 2 atomic force microscope (AFM) (Na-noscience Instruments, Inc., Phoenix, AZ) in dynamic mode withan ACL-A cantilever tip. Sample preparation for the AFMinvolved the deposition of our thin films onto silicon wafers.The AFM was used to characterize film roughness and unifor-mity. Cross-sections of the assemblies were imaged with a JEOL1200EXTEM(JEOLUSAInc., Peabody,MA)at anacceleratingvoltage of 100 kV. PS substrates were used instead of PET tofacilitate sectioning. After deposition, the film and substrate wereembedded in epoxy resin with a 1:1 anhydride/epoxide (A/E)ratio. This epoxy comprised Areldite 502 and Quetal 651 as theepoxy resin, along with a dodecenylsuccinic anhydride (DDSA)hardener andabenzyldimethylamine (BDMA)accelerator.Usingultramicrotomy, specimens were sectioned down to 70-110 nmthicknesses. These sections were vapor stained on nickel gridsusing aRuO4 staining solutionpreparedbyadding 1mLof 10w/v% sodium hypochlorite solution to 0.02 g of RuCl3.

56

Antimicrobial Effectiveness. The effectiveness of the anti-microbial films were tested using the Kirby-Bauer test, wherethe LbL-coated (or bare PET) disks are used in place ofantibiotic filter disks.57 The double membrane encapsulatedgram negative bacterium E. coli and the single membrane grampositive bacterium S. aureus were chosen as representatives ofthe major pathogenic groups for these tests. Gram positivebacteria are generally more susceptible to membrane perturb-ing agents such as CTAB than are gram negative bacteria.Luria broth and Luria broth plates were used throughout formicrobial growth. Films were deposited on PET substrates,with bare PET samples evaluated as controls. The zone ofinhibition (ZOI) of PET alone was zero, indicating that it is notinhibitory. Antimicrobial properties of various films weretested under a variety of conditions that include film composi-tion, number of bilayers, testing temperature, and age of agiven film. ZOI was recorded for each condition as the averageof 3 radial measurements, from the rim of the disk to thebeginning of bacterial growth, with 2 disks per condition, asshown in Figure 2.

Results and Discussion

Film Deposition and Microstructure. Figure 3a shows filmdeposition monitored using QCM. A control system (without theantimicrobial agent) is included to show the influence of CTABon growth. With the addition of CTAB, growth proceeds at amuch higher rate. Additionally, weight variation between PDDA+CTAB/PAA and the CTAB/PAA systems is minimal, suggest-ing that CTAB deposits to a much greater extent than PDDA.This is not surprising considering that CTAB is a much smallermolecule and likely has greater mobility in solution. Film thick-nesswasmeasured using profilometry at 7, 10, 15, and 20 bilayers,as shown in Figure 3b. Additionally, ellipsometry was performedto confirm thicknesses up to 5 bilayers and is included inSupporting Information. The growth trend here confirms thetrend obtained using QCM. Since the PDDA/PAA films exhibitmuch slower (thinner) growth, they were too thin for measure-ment using profilometry at less than 20BLandwere not analyzed.Again, film growth in the CTAB/PAA system was greater thanPDDA+CTAB/PAA. It has been shown that the addition ofsalts to LbL solutions yields much thicker films.58 In this system,CTAB is a salt, increasing ionic strength and screening charges onthe polymers. Similarly, when PDDA is removed from thecationic solution, the charge density decreases, and the resultingfilm is slightly thicker. In the absence of PDDA, rougher filmswith larger domain structure are generated. It is also possible thatCTAB molecules, consisting of a 16-carbon tail and cationicammonium headgroup, deposit as something resembling a lipidbilayer found in cell walls (shown schematically in Figure 1).59

This would account for the ability of singly charged CTAB togenerate the charge inversionnecessary to grow in the absence of ahighly charged cation such as PDDA. The change in growth slopearound the third bilayer is indicative of two growth modes:initially, islandic expansion followed by the typical vertical layeradsorption.60 In films with only single ingredients in each layer,film composition in weight or mole percent can be determinedfrom QCM data since this method obtains the weight of eachlayer deposited. For this analysis, CTAB molecular weight wascalculated without bromide because this is removed in aqueussolution. In the case of PAA, repeat unit molecular weight wasused. Calculations revealed that CTAB/PAA films are 20.1 mol%( 0.86 mol % CTAB and 79.9 mol % ( 0.86 mol % PAA.

The surfaces of the films were analyzed using atomic forcemicroscopy, as shown in Figure 4. Comparison of CTAB+

Figure 2. Result of a Kirby-Bauer test, in which PET diskscovered with a given assembly are placed on a bacteria-swabbedagar plate and incubated. The resulting ring of no bacterial growthis the zone of inhibition, which is the measure of antimicrobialefficacy.

(54) Owens, D. K. J. Appl. Polym. Sci. 1975, 19, 265–271.(55) Zhang, D.; Sun, Q.; Wadsworth, L. C. Polym. Eng. Sci. 1998, 38, 965–970.(56) Brown, G. M.; Butler, J. H. Polymer 1997, 38, 3937–3945.(57) Benson, H. J. Microbiological Applications: A Laboratory Manual in

General Microbiology, 3rd ed.; Wm. C. Brown Company Publishers: Duboque,IA, 1980.

(58) Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120, 7626–7634.(59) Becker, W. M.; Kleinsmith, L. J.; Hardin, J. The World of the cell, 6th ed.;

Pearson Education, Inc., publishing as Benjamin Cummings: San Fransisco, CA,2006.

(60) Ostrander, J. W.; Mamedov, A. A.; Kotov, N. A. J. Am. Chem. Soc. 2001,123, 1101–1110.

DOI: 10.1021/la901161z 10325Langmuir 2009, 25(17), 10322–10328

Dvoracek et al. Article

PDDA/PAA and CTAB/PAA films reveals a definite structuraldifference. The system without the polymer in the cationic layerhas a much rougher surface. The range of surface height is halvedwith the inclusion of PDDA. Also worth noting is that thisvariation in surface height seen in the CTAB/PAA system is onthe order of the overall surface height seen previously usingprofilometry. This data suggests a higher amount of CTABaggregation without PDDA, revealing better dispersion withthe addition of the polymer to the cationic layer. Attempts touse infrared microscopy were inconclusive, but this was expectedbecause the IR spot size of 10-15 μm does not provide highenough resolution to distinguish the regions in these films. Thecorresponding phase images, which are more akin to frictionalcontrast, do an even better job of highlighting the order inducedby the presence of PDDA in the assembly (Figure 4d). Phase

contrast is much less pronounced in the surface image of theassembly without PDDA (Figure 4c), which suggests there is lessphysical/chemical distinction between the bumps and the regionsbetween them.

Figure 5 showsTEMcross-sections of 10BL filmsdeposited onpolystyrene, with and without PDDA in the cationic layer. Itshould be noted that these images do not represent the full filmthickness because of low film strength that results in fractureduring sectioning. The films in these micrographs have a mottledappearance, indicating that the layers of the film intertwine anddiffuse among each other rather than laying down discretely.These high levels of diffusion during film deposition suggest thatCTAB will easily diffuse through the film during use, increasingantimicrobial efficacy. Additionally, the film created with PDDAin the cationic layer (Figure 5b) shows better uniformity because

Figure 3. Filmmass as a function of the number of layers deposited, asmeasuredwithQCM (a). Profilometrymeasurements are also shownto confirm the linear growth trend suggested by QCM (b).

Figure 4. AFMheight images of CTAB/PAA (a) and PDDA+CTAB/PAA (b) 10BL films deposited on silicon wafers. The correspondingphase images (c,d) for each of these systems are directly beneath the height images.

10326 DOI: 10.1021/la901161z Langmuir 2009, 25(17), 10322–10328

Article Dvoracek et al.

the PDDA spatially separates CTAB during deposition. Whilethis does not seem to affect initial antimicrobial activity, this mayincrease film longevity because diffusion out of the film involvesthe breakup of smaller aggregates. This affects film longevity,consequently affecting reliability, because diffusion out of the filmis less sporadic. It is these nano/microstructural characteristicsthat influence the antimicrobial behavior of these films, asdescribed in the next section.Antimicrobial Efficacy. All antimicrobial testing was done

using the Kirby-Bauer protocol.57 Films were deposited on PETsubstrates, with bare PET samples evaluated as controls. In boththe E. coli and S. aureus tests, the ZOI of the PET alone was zero,indicating that it is not inhibitory. Antimicrobial properties ofvarious filmswere testedunder a variety of conditions that includefilm composition, number of bilayers, testing temperature, andage of a given film.

The antimicrobial effectiveness of both PDDA+CTAB/PAAand CTAB/PAA were evaluated with both 10 and 20 bilayers ofdeposition. At 10 BL, films with PDDA exhibited a greater ZOI(i.e., greater antimicrobial efficacy) thanwithout. The results weresimilar at 20 BL, as shown in Figure 6. As was discussed in thepreceding section, PDDA in the cationic layer creates improved

dispersion of CTAB in solution. With improved dispersion ofCTAB molecules, the antimicrobial range is also increased. It islikely that the presence of PDDAallowsmoreCTAB to diffuse outof a given assembly, effectivelymaking the film behave as though ithas more biocidal agent. This is why films show increasing efficacybeyond 10 BL of deposition (i.e., more bilayers mean moreantimicrobials capable of diffusing out of the film). At this testingtemperature (37 �C), the maximum ZOI observed is approxi-mately 2.3 mm in the case of S. aureus. Increases in ancimicrobialefficacy above this point are not observed because of insufficienttime for antimicrobial diffusion prior to microbial growth.

It is important to see whether or not the antimicrobial agentdiffuses out of merely the top bilayer or if CTAB in the lowerlayers of the film diffuse out as well. By testing various films withdifferent numbers of bilayers, the ability of the bottom layers tocontribute to the overall antimicrobial actionwas tested, as shownin Figure 7. Equal amounts of CTAB were deposited in each ofthe 9.5 (9 bilayers plus one extra cationic antimicrobial layer) and10 bilayer films, and the antimicrobial effects are similar. Addi-tionally, S. aureus ZOI levels off very quickly with number ofbilayers, whereas E. coli seems to show increasing ZOI. This maybe due to the fact that S. aureus is more sensitive to CTAB andtherefore more quickly becomes governed by diffusion ratherthan release concentration. To further investigate the abilities ofCTAB to diffuse through the film, a 10 BL film was constructedwith no antimicrobial in the top five bilayers (5BL CTAB+PDDA/PAA followed by 5BL PDDA/PAA). These results,shown in Figure 8, demonstrate that CTAB diffuses throughmultiple layers. In fact, a comparison of the 5 BL film and the 10BL film with CTAB only in the lower five bilayers showsequivalent results within error. These results show that the PAAlayers do not hinder CTAB diffusion. PAA simply acts to providecharge inversion for electrostatic deposition (Figure 3).

At lower temperatures, bacteria grows more slowly, allowingCTAB more time to diffuse out into the test plates during theperiod of bacteria proliferation. In addition to body temperature(37 �C), antimicrobial testing was performed at temperatures of23 and 18.2 �C. At decreasing temperatures, bacterial growthrates are reduced, resulting in the larger ZOI observed with bothE. coli and S. aureus, as shown in Figure 9, due to the longer timeallowed for CTAB to diffuse. These data suggest that the reasonthese films do not experience ZOIs greater than 2.5 mm at bodytemperature is because the antimicrobial cannot travel furtherthan this distance during the bacterial growth period. Once

Figure 5. TEM cross-sections of CTAB/PAA (a) and PDDA+CTAB/PAA (b) 10 BL films on polystyrene.

Figure 6. Zone of inhibition for 10- and 20-bilayer films with orwithout PDDA in the cationic layers. Additional bilayers do notenhance PDDA+CTAB/PAA efficacy, but they do increase theefficacy of CTAB/PAA films.

Figure 7. Zone of inhibition as a function of the number ofbilayers of PDDA+CTAB/PAA. Error bars reflect maximiumand minimum zones of inhibition.

DOI: 10.1021/la901161z 10327Langmuir 2009, 25(17), 10322–10328

Dvoracek et al. Article

bacteria have established themselves, there will be no subsequentloss of density.Time Delay. The duration of the film efficacy was examined

by storing the films in a desiccator for varying lengths of timebefore testing, as shown in Figure 10. These stored films showeddecreased effectiveness initially, but the films maintained signifi-cant antimicrobial efficacy over the course of four weeks. It ispossible that these films rearrange to some equilibrium state afterdeposition. While antimicrobial activity may not be lost, someCTABmolecules may complex, decreasing their ability to diffuseout. This would explain the initial decrease and eventual levelingof efficacy. It seems that these films can be stored in dryenvironments for long periods of time without significant lossof antimicrobial efficacy.

For antimicrobial films, it is important to know how long filmswill remain active once inuse. In this case, duration of efficacywastested by performing these Kirby-Bauer tests overmultiple days,where disks were exposed to the moist environment of an agarplate in the incubater for a varying number of days prior toexposure to bacteria. Each day, the antimicrobial disks wereremoved and placed onto newly swabbed plates. New, clean disks

were exposed to bacterial growth each day, but all disks were cutfrom the same initial film and exposed to agar media withoutbacteria prior to testing. This use of clean disks for each dayensured that daily removal from the test plate would not influencethe results. These 10 BL films of PDDA+CTAB/PAA releasedCTAB strongly over two days, as shown in Figure 11. Resultsfrom this test demonstrate a decreasing effectiveness, with loss ofreliability after five days for samples with PDDA in the cationiclayer and after four days for samples without. Samples with onlyCTAB in the cationic layer (i.e., no PDDA) show some anti-microbial action at longer times. Since PDDAacts as a dispersingagent, as seen in both TEM micrographs (Figure 5) and AFMimages (Figure 4), discrepancies with films not containing PDDAmay be due to a somewhat variable CTAB concentration fromdisk to disk (i.e., PDDA-containing films are more homogeneousand consistent).

Conclusions

The introduction of antimicrobial agents into LbL filmsallows them to exhibit antimicrobial behavior. Additionally, this

Figure 9. Zone of inhibition as a function of temperature for 10BL PDDA+CTAB/PAA films. The larger ZOI at lower tempera-ture is attributed to slower bacterial growth and longer time forCTAB diffusion. Error bars reflect maximum and minimum ZOI.

Figure 10. Zone of inhibition as a function of storage time for 10BL PDDA+CTAB/PAA films. Prior to testing, films were storedin a dry environment. Error bars reflect ZOI.

Figure 11. Zone of inhibition as a function of days of exposure tothe Luria Broth plates (used for Kirby-Bauer testing) for 10 BLfilms. Both PDDA+CTAB/PAA and CTAB/PAA films wereevaluated to determine how long antimicrobial release will besustained when in use. Error bars reflect maximum and minimumZOI.

Figure 8. Zone of inhibition toE. coli and S. aureus for assembliesof varying composition.Theability ofCTABtodiffuse through thesystem was evaluated by building a film with CTAB only in thelower 5 BL using a PDDA+CTAB/PAA film. Error bars reflectmaximium and minimum zones of inhibition.

10328 DOI: 10.1021/la901161z Langmuir 2009, 25(17), 10322–10328

Article Dvoracek et al.

work shows that the polycation is not necessary to build theantimicrobial films, but it does improve film uniformity. Thelower uniformity of films without PDDA results in decreased filmhomogeneity and antimicrobial efficacy. A key learning from thiswork was that the CTAB molecules easily diffuse through theentireLbL film thickness.Using PDDA-PAAas a base forCTABincorporation produced the highest thickness growth rate, sug-gesting looser polymer packing that facilitates molecular move-ment through the system. Comparison of film activity at varyingtemperatures demonstrated higher bacterial killing abilities atlower temperatures where CTAB had longer time to diffuse outinto the system before bacterial growth. These observations, incombination with longevity studies, show that the best films areeffective over a 4-6 day period of activity upon continuous

exposure to healthy bacteria. This study provides a method toimprove antimicrobial efficacy and to use LbL assembly toproduce films with other types of biological activity (e.g., drugdelivery or enzyme stability).

Acknowledgment. We acknowledge financial support forthis work from the National Science Foundation GraduateResearch Fellowship and Texas Engineering Experiment Station(TEES).

Supporting Information Available: Film growth of threedifferent systems of varying polyelectrolyte strength combi-nations. This material is available free of charge via theInternet at http://pubs.acs.org.