Support Layer Influencing Sticking Probability: Enhancement ofMercury Sorption Capacity of GoldYlias M. Sabri, Samuel J. Ippolito,* and Suresh K. Bhargava*

Centre for Advanced Materials & Industrial Chemistry (CAMIC), School of Applied Sciences, RMIT University, Melbourne, VIC3001 Australia

*S Supporting Information

ABSTRACT: Elemental mercury (Hg0) sorption on Au thinfilms deposited over SiO2 (SiO2:Au) and Ti (Ti:Au) supportlayer is investigated by quartz crystal microbalance (QCM)technique. The Hg0 sorption capacity of Au thin films is shownto be greatly influenced by the underlying support layer, wherean enhancement of 267% was observed for SiO2:Au overTi:Au films for the same Au film thickness. Furthermore, byincreasing the Au film thickness, the total Hg0 captured wasobserved to increase even though the mercury sorptionefficiency (or mass ratio of Hg0 to Au) was observed todecrease for a 8 h Hg0 exposure period. Sticking probability calculations showed that the SiO2:Au substrate maintains a stickingprobability of ∼10−6, which is at least 2 orders of magnitude greater than that of the Ti:Au surface, thus indicating that Hg0

sorption capacity of a material can be greatly influenced by the choice of underlying support layer.

1. INTRODUCTION

Among various pollutants, mercury (Hg) has a significantimpact on many biological processes and the environment ingeneral. Elemental mercury (Hg0) represents 90−99% ofatmospheric mercury emissions, which is primarily emittedfrom burning coal and other industrial process. Due to its highvapor pressure and low solubility in water, gas phase Hg0 isdifficult to remove using conventional pollution controlscrubbing systems.1−4 Consequently, the lack of an environ-mentally friendly and economically viable mercury controlprocess has led to the closure of several cement plants withinthe USA and Europe.5−7 A current challenge faced by industryis the availability of suitable sorbent materials which meetrigorous requirements (being highly recyclable and cost-effective) to capture Hg0 from industry effluent streams.8−10

Currently, the most widely used process for removing Hg0 fromindustry stack gases is by activated carbon (AC) adsorption,metal oxides loaded AC, and metal/metal oxides supported onceramic and silicate particles.11−13 It has been estimated thatthe cost involved to treat a Hg0 contaminated stream withcarbon based sorbents can exceed US$50,000 per kg of Hgremoved.14 The main downside to using such processes is therequirement to minimize moisture build-up from the gas streamon the sorbent material and the fact that the bed can pose aserious fire hazard due to the exothermic adsorption of organiccompounds overtime. Furthermore, once the sorbent materialhas expired it must be relegated as toxic waste, furtherburdening landfill sites. In order to overcome these issues,metal and metal oxides on different supports developed byADA Technologies for example have been used in gas,petroleum, and alumina industries as they have been shown

to be efficient Hg0 scavengers;15−17 however, the uptake of suchmaterials has been limited. In this work we describe thedramatic influence the underlying support material has on theHg−Au sticking probability and hence Hg0 sorption capacity ofAu ultrathin film coatings on both SiO2 and Ti supportmaterials. By understanding the effect the underlying supportmaterial, it is proposed that better Hg0 scavenging efficienciescan be obtained from materials used for removing Hg0 from gasphase industry stack effluents or within laboratory basedpreconcentration equipment. The use of Au based sorptionmaterials would also have the added advantage of potentiallybeing significantly more economical than conventional sorptionbased processes as it can be regenerated to a certain extentfollowing mere dry nitrogen exposure18 or fully regeneratedusing heat treatment processes,19−21 thus reducing landfillrequirements for conventional spent sorbent based materialswhich either have limited, or no regenerable capabilities.Additionally the effect a support material is also studied inrespect to the Hg0 sorption capacity as a function of Au filmthickness in real-time. Monitoring of the Hg−Au interactionprocess was performed by fabricating quartz crystal micro-balance (QCM) with Ti and SiO2 supported Au thin films ofdifferent thickness. The QCM measurements made it possibleto measure the Hg0 capacity of each film during the Hg0

exposure process in real-time by utilizing the linear relationship(eq 1) between the frequency shift (Δf) and the mass change(Δm) observed on the electrode surface given by

Received: January 29, 2013Revised: March 18, 2013Published: March 19, 2013

Article

pubs.acs.org/JPCC

© 2013 American Chemical Society 8269 dx.doi.org/10.1021/jp4009724 | J. Phys. Chem. C 2013, 117, 8269−8275

μ ρΔ =

− Δ⇒ =f

S mn

Sf

A

2

q q

ff

02

(1)

where Sf represents the integral mass sensitivity or Sauerbreyconstant and is proportional to the square of the fundamentalfrequency, f 0, inversely proportional to the surface area, A, andincreases proportionally with the overtone number, n.22,23 Theparameters μq and ρq are the shear modulus (2.947 × 1011 g/cm s2) and density (2.648 g/cm3) of the quartz substrate.

2. EXPERIMENTAL SECTION2.1. QCM Fabrication. In order to study the Hg0 sorption

capacity of the Ti and SiO2 supported Au thin films of differentthickness, specially prepared QCMs (10 MHz) with multi-layered electrodes were fabricated as shown in Figure 1. Eight

separate QCMs devices, each having an Au sensitive layer ofeither 10, 20, 30, or 40 nm thick were fabricated. These four Aufilm thicknesses were chosen in-order to maintain thenanostructured topography of the upper Au sensitive layerused in this study. A minimum Au layer thickness of 10 nm waschosen as it has been previously reported that Au filmthicknesses above 7.5 nm is required to form a continuous filmcoverage.24 Four QCM devices employed a 20 nm thick Tilayer underneath the Au sensitive layer, whereas the other fourQCMs employed a 20 nm SiO2 layer. Similar to all devices,each of the above layer combinations were deposited over acommon QCM electrode configuration having a 150 nm thickAu electrode over a 10 nm Ti adhesion layer, which ensuredthat the Q factor of each device was maintained above 7000according to previous studies.25 Each QCM was fabricated sothe conductivity of the upper SiO2:Au or Ti:Au layers did notcontribute to the device’s performance. This was achieved byusing a specially designed shadow mask to stop any overlapbetween the upper layer and the primary underlying 150 nm Aufilm electrodes of the QCM. All depositions were performed ata rate of 0.2 nm/s on both faces of the quartz crystals by aBalzers e-beam (BAK 600) evaporator at room temperature(∼22 °C). A base pressure of 1.5 × 10−7 Torr was used prior todeposition. Both the Ti and SiO2 layers on the Au electrode ofthe quartz crystal substrates were coated using the samedeposition conditions so that the morphological characteristicsof the outermost Au film was only influenced by the underlyingsupport layer material.

2.2. Hg0 Exposure and Monitoring. The same exper-imental setup described in our prior work was used, details ofwhich can be found elsewhere.26,27 Briefly, mercury vaporconcentrations were achieved by using mercury permeationtubes which were purchased from VICI, TX, USA. DifferentHg0 concentrations were generated using a PID temperaturecontroller to heat the permeation tubes at different set points.The Hg0 concentration generated at each set point wascalibrated using an acidic KMnO4 wet trapping method, whichwas adapted from the Ontario Hydro method.26,28 The Hg0

trapped in the KMnO4/H2SO4 solution was quantified usingAgilent Technologies HP7700 series inductively coupledplasma mass spectroscopy.The change in resonant frequency (Δf) of the QCMs was

measured using a Maxtek RQCM (10 MHz phase lockedoscillator, USA) with resolution of ±0.01 Hz. The QCMs wereexposed to dry N2 for a 100 min period while monitoring itsresponse in order to ensure the devices were stable prior to anyHg0 exposure experiments. The QCMs were then tested towardthree Hg0 concentrations of 3.65, 5.70, and 10.55 ± 0.05 mg/m3 (balance N2) using 8 h exposure times for each Hg0

concentration followed by a 5 h recovery under dry N2atmosphere. The purpose of the 5 h recovery period was toremove any loosely attached Hg0 atoms from the surface of theAu thin film based QCM prior to removing the device from thegas chamber. A freshly made, unexposed QCM was used foreach Hg0 exposure experiment. The total gas flow rate into thechamber was kept constant at 200 sccm using a custom builtfour channel mass flow controller system. A constant pressureof ∼1200 Torr was maintained throughout the entireexperiment. The approximate volume of the chamber housingthe QCM sensors was ∼0.1 L.

2.3. Surface Characterization. Scanning electron micro-scope (SEM) characterization was performed using a FEI NovaNanoSEM instrument operating at an accelerating voltage of 10kV, spot size of 3 and working distance of 4.5 mm. Atomicforce microscopy (AFM) tapping mode measurements wereperformed on the relevant films using a NanoScope IIIaMultimode AFM (Vecco, USA) under air-ambient conditions(20 °C and 40% RH). Rotated Monolithic silicon probes(TAP300-G) with a symmetric tip shape, a resonant frequencyof 300 kHz, a spring constant of 40 N m−1, and a radius ofcurvature of tip <10 nm (uncoated) were purchased fromBudget Sensors. Minor scan line errors were removed usingAFM software. More than 5 AFM scans were carried out ondifferent locations for each sample, and the results presentedwithin this manuscript are typical examples. The AFM scananalysis software was provided by Vecco Co. (USA) with theAFM instrument and was used to derive the surface analysiscalculations.

3. RESULTS AND DISCUSSION3.1. Surface Morphology. In order to determine the

influence of the underling Ti or SiO2 support on the upper Ausensitive layers’ topology and roughness, AFM characterizationof both support layers with each having an overlayer Au film of40 nm thickness was preformed, the results of which are shownin Figure 2. The average roughness (Ra), the root-mean-squareroughness (Rq), and surface areas (SA) of the scannedgeometric surface area (1 μm2) are observed to be similar forboth 40 nm thick Au films. The surface area of each sample isalso observed to have changed only 2−3% due to themorphology and roughness of the respective support layers.

Figure 1. Ultrathin Au sensitive layers deposited on quartz substrates.The outermost layer is comprised of the ultrathin films of Au (10−40nm Au sensitive layer) deposited on either a (a) Ti or (b) SiO2support layer. The inner Au layer is the QCM electrode having athickness of 150 nm.

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp4009724 | J. Phys. Chem. C 2013, 117, 8269−82758270

Phase mapping of the same areas show a clear distinctionbetween the two Au films indicating that their support layer hasaffected the topology of the Au film. Further evidence that theAu layer formed on the SiO2 support has large surface graindensity with sizes less than 20 nm as well as surface voids issupported by the SEM characterization of the same surfaces(see the Supporting Information, Figure S1). The SEMcharacterization also shows that the Au layer on the Tisupporting layer forms a uniform surface with larger Au grainsizes of ∼50 nm compared to the <20 nm grains observed onthe Au film with the SiO2 support.

3.2. Hg0 Sorption Capacity. 3.2.1. Influence of SupportLayer. Following the QCM fabrication and characterizationprocesses, the QCMs were exposed to Hg0 vapor concen-trations of 10.55 mg/m3 at an operating temperature of 28 °Cin order to determine the Hg0 sorption capacity of the Au thinfilm combinations. The Hg0 exposure period was fixed at 8 hfollowed by a 5 h recovery period under dry nitrogen. Aconstant gas flow of 200 sccm was maintained throughout theexpermiments. Figure 3a shows the Hg0 sorption capacity of a40 nm Au film on both the Ti and SiO2 underlying layers. Itwas observed that the Hg0 sorption capacity of the Au thin films

Figure 2. AFM images showing 40 nm thick Au film deposited on 20 nm SiO2 support layer (a) with the corresponding phase image (b) and (c) 20nm Ti support layer with the corresponding phase image (d). Scale for x, y, and z axis are 200, 200, and 20 nm per division, respectively. Thescanned geometrical surface area was 1 μm2 for all images shown.

Figure 3. QCM response showing the Hg0 sorption capacity of (a) Au films having a thickness of 40 nm deposited on either a 20 nm SiO2 or Tisupport layer and (b) Au films having a thickness of 10, 20, and 30 nm with SiO2 as the support material, exposed to Hg

0 concentration of 10.55 mg/m3 for 8 h followed by 5 h of dry N2 exposure at an operating temperature of 28 °C.

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp4009724 | J. Phys. Chem. C 2013, 117, 8269−82758271

using the SiO2 support layer was enhanced by 267% over thefilm with the Ti support layer. This was in spite of the AFMmeasurements in Figure 2 showing similar roughness andsurface area between the two Au films with different supportmaterials. However, phase mapping of the same areas show aclear distinction between the two Au films indicating that theirsupport layer has indeed affected the topology of the Au film,which is thought to be the primary contributing factor to thevastly different Hg0 sorption capacity of each film. In addition,both Ti and SiO2 is known to have little affinity towardHg09,16,27 however the combination of SiO2:Au surface isobserved to have an enhanced Hg0 sorption capacity over theTi:Au surface. By analyzing the rate (slope) of Hg0 sorption inFigure 3a for t greater than 200 min, it is observed that the rateis higher for the SiO2 support, indicating that the Au film is amore efficient Hg0 scavenger when deposited on SiO2 over Tisupport layer. As we have previously shown that Hg0 nucleatesbetter on surfaces with a higher density of grain boundaries andsurface defects,29 we postulate that the SiO2:Au combination isrich in these types of Hg0 nucleation sites. Several possibleamalgam formation mechanisms could occur when Hg0

sorption occurs on these nucleation sites; however, from theAu−Hg phase diagram the main amalgam expected on thesurfaces in this study is the Au2Hg amalgam26 which is the moststable phase below temperatures of 122 ◦C.By further analyzing Figure 3a, it can be observed that,

during the 5 h recovery period (t > 600 min), the SiO2:Au andTi:Au films desorbs significant amount (∼45 and ∼42%,respectively) of the adsorbed Hg0. This clearly demonstratesthat it is possible to regenerate the surface without the need forany heat treatment processes. Complete regeneration of thesurface would be further possible if the surfaces were elevatedto above 150 °C, as it is widely recognized that Au films freelydesorb or release captured mercury when heat treated to above150 °C.29−31

3.2.2. Influence of Au Film Thickness. In order to study theeffect of Au film thickness on the SiO2 support, QCMscontaining 10, 20, and 30 nm Au films as the final layer werealso exposed to Hg0 vapor under the same conditions, theresults of which is shown in Figure 3b. It can be observed thatas the Au film thickness is increased, the Hg0 sorption capacityof the Au film also appears to increase, however in a nonlinearmanner. By doubling, tripling or quadrupling the Au filmthickness from 10 to 20 or 30 or 40 nm, the sorption capacityfollowing 8 h of Hg0 exposure was observed to increase by

∼80%, ∼133% and ∼165%, respectively. The recovery (ratio ofHg0 desorbed to total Hg0 originally available on the surface) ofeach Au film was also found to range from 45 to 50%, thusindicating that Au film thickness does not significantly influencethe Hg desorption processes for the tested Au film thickness.However, overall the 10 nm Au film was found to have thehighest mercury sorption capacity per nanometer of Au filmthickness deposited over the SiO2 layers, which correlated to a22 ng/cm2 of Hg0 sorption capacity (or 7.3 ngHg0/ngAu) overthe 8 h Hg0 sorption period. This indicates that the increasedamount of Au deposited over the SiO2 support would notnecessarily increase its Hg0 sorption capacity of the surface.

3.2.3. Influence of Operating Temperature. For practicalreasons, since most industrial applications typically operate atelevated temperatures,32 the SiO2:Au and Ti:Au films weretested for their Hg0 sorption capacity at an operatingtemperature of 90 ± 1 °C. All other experimental andoperating conditions were kept the same as in sections 3.2.1and 3.2.2. Figure 4a shows the Hg0 sorption capacity which wasderived from the QCM response data of the 10, 20, 30, and 40nm Au films deposited on a 20 nm SiO2 support layer. It can beobserved that the 10 nm Au film has the most efficient Hg0

sorption kinetic where the amount of captured Hg0 is as high asthe 30 nm Au film at the end of the 8 h Hg0 exposure period.This indicates that the 10 nm Au film over a SiO2 layer wouldbe the best material combination for use as an efficient Hg0

scavenging surface. Similar kinetics were also observed forsample exposed to lower Hg0 concentrations (<10.55 mg/m3),as shown in Figure S2 of the Supporting Information. Similarly,it was observed that all of the SiO2:Au films desorbapproximately 50−55% of the adsorbed Hg0, which is a slightlyhigher recovery rate than that observed rate shown for the samesurface in Figure 3a when operating at 28 °C. However, it isworth noting that the total Hg0 sorption capacity of the 30 nmsampled reduces with increasing operating temperature,undergoing a 30% lower mercury sorption capacity at anoperating temperature of 90 °C compared to that at 28 °C.In order to determine if further increases in the Au film

thickness can be used to increase the sorption capacity of thesurface, the Hg0 sorption capacity of Au films having a filmthicknesses of 50, 100, 150, and 200 nm deposited Au over a 10nm thin Ti adhesion layer were also assessed at 90 °C, as shownin Figure 4b. It is again observed that Au films deposited on Tisupport have less Hg0 sorption capacity than those depositedon SiO2 support even though much thicker Au films were used

Figure 4. QCM response showing the Hg0 sorption capacity of Au films (a) having thicknesses of 10, 20, 30, and 40 nm with SiO2 as the supportmaterial and (b) having thicknesses of 50, 100, 150, and 200 nm with Ti as the support material, exposed to Hg0 concentration of 10.55 mg/m3 for 8h at an operating temperature of 90 °C.

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp4009724 | J. Phys. Chem. C 2013, 117, 8269−82758272

when using Ti as the support material. Furthermore, It isobserved that the Hg0 sorption capacity of the Au filmsdeposited on Ti support does not increase by increasing thefilm thickness >150 nm following an 8 h Hg0 exposure periodat 90 °C. The amount of Hg0 that was released from the surfacefollowing 5 h of dry N2 gas exposure was found to be ∼70% ofthe adsorbed Hg0 for all four film thicknesses. The resultsindicate that the Hg0 sorption capacity of Au (and potentiallyother materials) can be enhanced by choosing the right supportmaterial and film thickness. Subsequently the data also showsthat sorption capacity does not necessarily increase by simplyincorporating more of the sorption material on a supportmaterial.3.3. Hg0 Sorption Model. A better understanding of Hg0

sorption on Au films deposited on either Ti or SiO2 adhesionlayers may be achieved if the Hg0 sorption rates (k) andsticking probability are calculated using the QCM data. In orderto do this, the Hg0 sorption kinetics needs to be modeled as amathematical function which would fit the QCM response datafor all Au films deposited on different support materials. Wehave previously29 shown that for a 60 min Hg0 exposure periodthe sorption kinetics were best described by the rationalfunction. However in this study, the Hg0 sorption kinetics overan 8 h period is found to be better described by an exponentialgrowth model (eq 2)

= + +n t a a d( ) e et b t cHg 1

/2

/(2)

with R2 > 0.995 as shown in Figure 5. This type of model wasthe best fit for all tested operating temperatures and Hg0

concentrations. Here, (nHg(t)) represents the amount of Hg0

undergoing sorption on the Au film during Hg0 exposure andthe constants a1, a2, b, c, and d are dependent on temperatureand Hg0 concentration. The constants a1 and a2 indicate themagnitude of the role each exponential component has on theoverall model, whereas the constants b and c determine howquickly that effect deteriorates with time. The constant ddetermines the maximum amount of mercury that can undergosorption on the surface.The rate k(t) at which Hg0 undergoes sorption on the Au

surface is evaluated as in eq 3 and derived from eq 2

= +k tab

ac

( ) e et b t c1 / 2 /(3)

The fact that the summation of two exponential terms bestdescribes the Hg0 sorption rate suggests that there are at least

two processes that occur when Hg0 is exposed to an Au thinfilm. The two processes are postulated to be Hg0 adsorption onthe Au surface and Hg0 diffusion through the Au surface. Thisfinding also agrees well with our previous work29 where wehave shown that part of the QCM response curve is attributedto the Hg diffusion process which occurs when the Au film isfirst exposed to Hg0 vapor.

3.4. Hg−Au Sticking Probability. In order to observe andcompare the affinity of the SiO2:Au and Ti:Au surfaces towardHg0 vapor, the sticking probability (S) of Hg−Au for eachsurface was calculated from the QCM data obtained over the 8h Hg0 exposure period at an operating temperature of 28 °C.Sticking probability is defined as the ratio of the rate of sorption(Kads or k(t)), defined in eq 3 to the rate at which Hg0 atomsfrom the vapor phase strike the Au surface (ZW).

33 The Sprofiles of both Au surfaces when exposed to Hg0 vaporconcentration of 10.55 mg/m3 are shown in Figure 6. It can be

observed that the sticking probability of Ti:Au reducessignificantly reaching from 1.0 to a value of ∼10−8 which is 2orders of magnitude lower than that of the SiO2:Au surface.The vast reduction in S for the Ti:Au may be due to a lack ofdiffusion of Hg into the surface. Therefore, it is highly likelythat the Hg diffuses right through the Au thin-film deposited onSiO2, thus allowing Hg0 sorption sites to become available formore Hg0 atoms to undergo sorption and diffusion processes atthe surface. That is, Hg diffusion processes is expected to bebetter favored on SiO2:Au over the Ti:Au surface. This is

Figure 5. QCM response data (dashed) and fitted (solid lines) using eq 3 during Hg sorption when the QCM was exposed to a Hg vaporconcentration of 10.55 mg/m3 at operating temperatures of 28 °C. The sensor response has been converted to mass of Hg vapor undergoingsorption per area (ng/cm2).

Figure 6. Sticking probability of Au films having a thickness of 40 nmdeposited on either a 20 nm SiO2 or Ti support layer exposed to Hg0

concentration of 10.55 mg/m3 for 8 h at an operating temperature of28 °C.

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp4009724 | J. Phys. Chem. C 2013, 117, 8269−82758273

because the Au on the SiO2 support is expected to be moremobile than on a Ti support due to the better adhesion of Auon Ti over the SiO2 support layer.

34,35 Therefore, the better Hgdiffusion in the SiO2:Au surface is expected to be due to thebetter mobility of the Au on the SiO2 as there was no evidenceof any porosity on the surface from our AFM and SEMcharacterization studies.

4. CONCLUSIONSUsing the QCM technique we have shown that the Hg0

sorption capacity of an Au thin film can be significantlyenhanced by choosing the correct support material, eventhough AFM measurements might show similar roughness andsurface area properties between the Au films regardless of theunderlying layer. The Hg0 sorption capacity of the Au filmswere shown to have increased up to 4-fold by using SiO2instead of Ti as the support material. The increase is attributedto the SiO2:Au surface maintaining more than 2 orders ofmagnitude higher Hg−Au sticking probability than the Ti:Ausurface during the Hg0 exposure period. The effect of Au filmthickness was also assessed to show that the Hg0 scavengingefficiency of the Au film increases with increasing Au filmthickness; however, the overall sorption capacity does notnecessarily increase by simply incorporating more sorptionmaterial (Au) on either the SiO2 or Ti support. Of the testedAu thin film combinations, the 10 nm Au film deposited over aSiO2 support layer was observed to be the most efficient Hg0

scavenger based on Hg0 absorbed per mass of Au used. It isenvisaged that, by understanding the influence of the supportlayer using this approach, the Hg0 sorption capacity andefficiency of materials such as Au and possibly other noblemetals can be enhanced with the aim of reducing the requiredquantity of the sorbent layer whiles still maintaining a highsorption capacity properties.

■ ASSOCIATED CONTENT*S Supporting InformationFigure S1 - SEM images showing 40 nm thick Au filmdeposited on 20 nm SiO2 support layer (a) and 20 nm Tisupport layer (b). Scale bars represent 500 nm. The image in(a) shows much smaller grains and “cracks” on the surfacewhich could potentially have high number density of surfacedefects and therefore Hg0 sorption sites resulting in itsenhanced mercury sorption capacity. Figure S2 - QCMresponse showing the elemental mercury sorption capacity ofAu films having a thickness of 10, 20, 30, and 40 nm with SiO2as the support material, exposed to Hg0 concentrations of (a)3.65 mg/m3 and (b) 5.70 mg/m3 at and operating temperatureof 90 °C. This material is available free of charge via theInternet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: samuel.ippolito@rmit.edu.au; suresh.bhargava@rmit.edu.au. Phone: +61 3 9925 2330.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThe authors acknowledge the Australian Research Council(ARC) for supporting this project (Grant No. LP100200859)and the RMIT microscopy and microanalysis facility (RMMF)

for allowing the use of their comprehensive facilities andservices. S.J.I. acknowledges ARC for APDI fellowship.

■ REFERENCES(1) Qui, J. Tough talk over mercury treaty. Nature 2013, 493, 144−145.(2) Presto, A. A.; Granite, E. J. Survey of catalysts for oxidation ofmercury in flue gas. Environ. Sci. Technol. 2006, 40 (18), 5601−5609.(3) James, J. Z.; Lucas, D.; Koshland, C. P. Gold Nanoparticle FilmsAs Sensitive and Reusable Elemental Mercury Sensors. Environ. Sci.Technol. 2012, 46 (17), 9557−9562.(4) Ramesh, G. V.; Radhakrishnan, T. P. A Universal Sensor forMercury (Hg, HgI, HgII) Based on Silver Nanoparticle-EmbeddedPolymer Thin Film. ACS Appl. Mater. Interfaces 2011, 3 (4), 988−994.(5) Paone, P. In Mercury controls for the cement industry; CementIndustry Technical Conference, March 28, 2010; pp 1-12.(6) Impact of EPA’s Regulatory Assault on Power Plants: NewRegulations to Take 34 GW of Electricity Generation Offline and thePlant Closing Announcements Keep Coming; Institute for EnergyResearch: 2012; pp 1-40.(7) Sullivan, E. Overview Impact of Esisting and Proposed RegulatoryStandards on Domestic Cement Capacity. http://www.cement.org/Econ_Anal_%202-1-11.pdf (accessed 2012).(8) Reddy, B. M.; Durgasri, N.; Kumar, T. V.; Bhargava, S. K.Abatement of Gas-Phase MercuryRecent Developments. Catal. Rev.,Sci. Eng. 2012, 54 (3), 344−398.(9) Granite, E. J.; Pennline, H. W.; Hargis, R. A. Novel Sorbents forMercury Removal from Flue Gas. Ind. Eng. Chem. Res. 2000, 39 (4),1020−1029.(10) Granite, E. J.; Myers, C. R.; King, W. P.; Stanko, D. C.;Pennline, H. W. Sorbents for Mercury Capture from Fuel Gas withApplication to Gasification Systems. Ind. Eng. Chem. Res. 2006, 45(13), 4844−4848.(11) Mei, Z. J.; Shen, Z. M.; Zhao, Q. J.; Wang, W. H.; Zhang, Y. J.Removal and recovery of gas-phase element mercury by metal oxide-loaded activated carbon. J. Hazard. Mater. 2008, 152 (2), 721−729.(12) Fan, X.; Li, C.; Zeng, G.; Gao, Z.; Chen, L.; Zhang, W.; Gao, H.Removal of Gas-Phase Element Mercury by Activated Carbon FiberImpregnated with CeO2. Energy Fuels 2010, 24 (8), 4250−4254.(13) Yuan, C. S.; Lin, H. Y.; Wu, C. H.; Liu, M. H. Partition and sizedistribution of heavy metals in the flue gas from municipal solid wasteincinerators in Taiwan. Chemosphere 2005, 59 (1), 135−145.(14) Turchi, C. S. Novel Process for Removal and Recovery of VaporPhase Mercury; ADA Technologies, Inc.: USA, 2000; pp 1−57.(15) Yan, T. Y. Removal of mercury from natural gas and liquidhydrocarbons utilizing silver on alumina adsorbent. Patent number:5,053,209, 1989.(16) Granite, E. J.; Hargis, R. A.; Pennline, H. W. Sorbents for mercuryremoval from flue gas; Report Number: DOE/FETC/TR--98−01;Federal Energy Technology Center: Pittsburgh, 1998; p 47.(17) Roberts, D. L.; Albiston, J.; Broderick, T.; Greenwell, C.;Stewart, R. Novel Process for Removal and Recovery of Vapor PhaseMercury; ADA Technologies, Inc.: USA, 1998; p 92.(18) Sabri, Y. M.; Ippolito, S. J.; Atanacio, A. J.; Bansal, V.; Bhargava,S. K. Mercury vapor sensor enhancement by nanostructured golddeposited on nickel surfaces using galvanic replacement reactions. J.Mater. Chem. 2012, 22 (40), 21395−21404.(19) Morris, T.; Sun, J.; Szulczewski, G. Measurement of thechemical and morphological changes that occur on gold surfacesfollowing thermal desorption and acid dissolution of adsorbedmercury. Anal. Chim. Acta 2003, 496 (1−2), 279−287.(20) Fialkowski, M.; Grzeszczak, P.; Nowakowski, R.; Holyst, R.Absorption of Mercury in Gold Films and Its Further Desorption:Quantitative Morphological Study of the Surface Patterns. J. Phys.Chem. B 2004, 108 (16), 5026−5030.(21) Nowakowski, R.; Kobiela, T.; Wolfram, Z.; Dus, R. Atomic forcemicroscopy of Au/Hg alloy formation on thin Au films. Appl. Surf. Sci.1997, 115 (3), 217−231.

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp4009724 | J. Phys. Chem. C 2013, 117, 8269−82758274

(22) Janshoff, A.; Galla, H.-J.; Steinem, C. Piezoelectric Mass-SensingDevices as Biosensors - An Alternative to Optical Biosensors? Angew.Chem. 2000, 39 (22), 4004−4032.(23) Janshoff, A.; Steinem, C. Quartz Crystal Microbalance forBioanalytical Applications. Sensors Update 2001, 9 (1), 313−354.(24) Haskell, R. L. B.; Caron, J. J.; Duptisea, M. A.; Ouellette, J. J.;Vetelino, J. F. In Effects of film thickness on sensitivity of SAW mercurysensors; Ultrasonics Symposium, 1999. Proceedings, Orono, ME, IEEE:Orono, ME, 1999; pp 429−434, Vol. 1.(25) Sabri, Y. M.; Ippolito, S. J.; Al Kobaisi, M.; Griffin, M. J.; Nelson,D.; Bhargava, S. K. Investigation of Hg Sorption and DiffusionBehavior on Ultra-thin Films of Gold Using QCM Response Analysisand SIMS Depth Profiling. J. Mater. Chem. 2012, 22 (39), 20929−20935.(26) Sabri, Y. M.; Ippolito, S. J.; O’Mullane, A. P.; Tardio, J.; Bansal,V.; Bhargava, S. Creating gold nanoprisms directly on quartz crystalmicrobalance electrodes for mercury vapor sensing. Nanotechnology2011, 22 (30), 305501−09.(27) Sabri, Y. M.; Kojima, R.; Ippolito, S. J.; Wlodarski, W.; Kalantar-zadeh, K.; Kaner, R. B.; Bhargava, S. K. QCM Based Mercury VaporSensor modified with polypyrrole supported Palladium. Sens. Actuators,B 2011, 160 (1), 616−622.(28) Laudal, D.; Nott, B.; Brown, T.; Roberson, R. Mercuryspeciation methods for utility flue gas. Fresenius J. Anal. Chem. 1997,358 (3), 397−400.(29) Sabri, Y. M.; Ippolito, S. J.; Tardio, J.; Bhargava, S. K. Study ofSurface Morphology Effects on Hg Sorption-Desorption Kinetics onGold Thin-Films. J. Phys. Chem. C 2012, 116 (3), 2483−2492.(30) Scheide, E. P.; Taylor, J. K. Piezoelectric sensor for mercury inair. Environ. Sci. Technol. 1974, 8 (13), 1097−1099.(31) Scheide, E. P.; Taylor, J. K. A piezoelectric crystal dosimeter formonitoring mercury vapor in industrial atmospheres. Am. Ind. Hyg.Assoc. J. 1975, 36 (12), 897−901.(32) Levlin, M.; Niemi, H. E. M.; Hautojar̈vi, P.; Ikav̈alko, E.;Laitinen, T. Mercury adsorption on gold surfaces employed in thesampling and determination of vaporous mercury: a scanningtunneling microscopy study. Fresenius J. Anal. Chem. 1996, 355 (1),2−9.(33) Sabri, Y. M.; Ippolito, S. J.; Tardio, J.; Atanacio, A. J.; Sood, D.K.; Bhargava, S. K. Mercury diffusion in gold and silver thin filmelectrodes on quartz crystal microbalance sensors. Sens. Actuators, B2009, 137 (1), 246−252.(34) Sexton, B. A.; Feltis, B. N.; Davis, T. J. Characterisation of goldsurface plasmon resonance sensor substrates. Sens. Actuators, A 2008,141 (2), 471−475.(35) Homola, J.; Yee, S. S.; Gauglitz, G. Surface plasmon resonancesensors: review. Sens. Actuators, B 1999, 54 (1−2), 3−15.

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp4009724 | J. Phys. Chem. C 2013, 117, 8269−82758275