Environmental Effects on Zirconium Hydroxide Nanoparticles and

Subscriber access provided by READING UNIV

Article

Environmental Effects on Zirconium Hydroxide Nanoparticles and Chemical Warfare Agent Decomposition: Implications of Atmospheric Water and Carbon Dioxide Robert B. Balow, Jeffrey G. Lundin, Grant C. Daniels, Wesley O. Gordon, Monica McEntee, Gregory W. Peterson, James H. Wynne, and Pehr E Pehrsson ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b10902 • Publication Date (Web): 20 Oct 2017 Downloaded from http://pubs.acs.org on October 23, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Environmental Effects on Zirconium Hydroxide Nanoparticles and Chemical Warfare Agent Decomposition: Implications of Atmospheric Water and Carbon Dioxide Robert B. Balow,† Jeffrey G. Lundin,‡ Grant C. Daniels,‡ Wesley O. Gordon,§ Monica McEntee,§ Gregory W. Peterson,§ James H. Wynne,‡ and Pehr E. Pehrsson‡,* †

National Research Council Research Associateship Program, U.S. Naval Research Laboratory 4555

Overlook Avenue, SW, Washington, DC 20375, United States. ‡

U.S. Naval Research Laboratory, Chemistry Division, 4555 Overlook Avenue, SW, Washington, DC

20375, United States. §

U.S. Army, Edgewood Chemical Biological Center, 5183 Blackhawk Road, Aberdeen Proving Ground,

Maryland 21010, United States.

Keywords: Zr(OH)4, decontamination, DMMP, dimethyl methylphosphonate, FTIR, catalysis, operando, and spectroscopy.

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 32

ABSTRACT Zirconium hydroxide (Zr(OH)4) has excellent sorption properties and wide-ranging reactivity towards numerous types of chemical warfare agents (CWAs) and toxic industrial chemicals. Under pristine laboratory conditions, the effectiveness of Zr(OH)4 has been attributed to a combination of diverse surface hydroxyl species and defects; however, atmospheric components (e.g. CO2, H2O, etc.) and trace contaminants can form adsorbates with potentially detrimental impact to the chemical reactivity of Zr(OH)4. Here, we report the hydrolysis of a chemical warfare agent simulant, dimethyl methylphosphonate (DMMP) on Zr(OH)4 determined by gas chromatography mass spectrometry and in situ attenuated total reflectance Fourier transform infrared spectroscopy under ambient conditions. DMMP dosing on Zr(OH)4 formed methyl methylphosphonate (MMP) and methoxy degradation products on free bridging and terminal hydroxyl sites of Zr(OH)4 under all evaluated environmental conditions. CO2 dosing on Zr(OH)4 formed adsorbed (bi)carbonates and interfacial carbonate complexes with relative stability dependent on CO2 and H2O partial pressures. High concentrations of CO2 reduced DMMP decomposition kinetics by occupying Zr(OH)4 active sites with carbonaceous adsorbates. Elevated humidity promoted hydrolysis of adsorbed DMMP on Zr(OH)4 to produce methanol and regenerated free hydroxyl species. Hydrolysis of DMMP by Zr(OH)4 occurred under all conditions evaluated, demonstrating promise for chemical decontamination under diverse, real-world conditions. INTRODUCTION Much effort focuses on developing materials and sorbents for decontaminating toxic industrial chemicals (TICs)1 and chemical warfare agents (CWAs);2-3 however, the diverse reactivity and decomposition pathway of various TICs and CWAs makes finding an all-in-one decontamination product difficult.4 As a result, a wide array of materials, such as metal-organic frameworks (MOFs),5


2

Page 3 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


polyoxometalates,6 and high surface area nanoparticles and zeolites,7 have been investigated to address these challenges. One promising candidate, zirconium hydroxide (Zr(OH)4), shows excellent sorption and wide-ranging reactivity towards numerous TICs,8-10 CWAs, and CWA simulants.11-14 In a study by Bandosz et al., Zr(OH)4 showed nearly instantaneous decontamination of the toxic nerve agent O-ethyl S-[2-(diisopropylamino)ethyl] methylphosphonothioate] (VX).15 Peterson et al. demonstrated the removal of cyanogen chloride14 and sulfur dioxide11 TICs using Zr(OH)4. The reactivity of zirconium hydroxide has been attributed to a combination of diverse surface hydroxyl species (terminal, bridging, etc.), bonding configurations (monodentate, bidentate, bridging, etc.), and defect sites (such as under coordinated Zr), yielding both Brønsted and Lewis acid and base sites of varying strengths.11-13, 16 This reactivity led to the incorporation of Zr(OH)4 in a variety of composite materials and zeolites for sequestering and decomposing TICs and CWAs;8,

10

however, many of these promising preliminary

results were obtained without fully understanding the impact of atmospheric components (e.g., H2O and CO2) at ambient pressure. The disparity between a pristine laboratory environment and operationally relevant conditions can realize different outcomes during implementation of such technologies in the field, potentially leading to unexpected decomposition performance and complications with dependability and deployment. Many metal (hydr)oxides spontaneously react with H2O and CO2 to form surface bound species, which can hinder reactivity with CWAs or TICs by passivating or blocking reactive sites.17-20 More specifically, metal (hydr)oxides, including zirconia, rapidly form surface bound (bi)carbonate species when exposed to CO2.17-19, 21-23 Atmospheric water can exacerbate these surface reactions, as observed by Sorescu et al., who noted enhanced CO2 surface adsorption on wet TiO2 due to increased hydrogen bonding.24 Surface bound water also plays a significant role in surface reaction chemistry because it can alter reactant transport to the metal (hydr)oxide surface by forming additional interfaces.20 Gankanda et


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 32

al. observed varying CO2 reactivity with ZnO and CuO nanoparticle surfaces, depending on the relative humidity (RH) during CO2 exposure. At high RH, surface speciation changed from (bi)carbonates and carboxylates to water-solvated carbonates.25 On the other hand, water may enhance the reactivity of metal (hydr)oxides for decontaminating toxic chemicals by replenishing reactive surface hydroxyls depleted by hydrolysis, a common pathway for CWA decontamination.4 For these reasons, a deeper understanding of ambient surface-bound contaminants and Zr(OH)4 surface reactivity is critical for achieving robust and efficient chemistry for real-world decontamination challenges. In this work, the surface reactivity of Zr(OH)4 nanoparticles under ambient conditions with atmospheric components such as H2O and CO2 was investigated using time-resolved in situ attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) to identify adsorbed surface species and their relative stabilities. These surfaces were then challenged with dimethyl methylphosphonate (DMMP), a nerve agent analog, to understand the interplay between atmospheric adsorbates on Zr(OH)4 nanoparticles and DMMP. Gas chromatography mass spectrometry (GC-MS) were performed to quantify the relative impact of atmospheric components on Zr(OH)4 decomposition performance. Collectively, these data provide needed insight into interfacial and environmental chemistries occurring on Zr(OH)4 nanoparticles when exposed to a nerve agent simulant under operationally relevant conditions.

MATERIALS AND METHODS Reagents. All materials were used as received unless otherwise noted. Zr(OH)4 powder was obtained from Magnesium Elektron Ltd (MEL) with a surface area of approximately 407 m2/g calculated by Brunauer-Emmett-Teller analysis using nitrogen (N2) adsorption. Ultra-high purity (UHP) N2 (99.999%,) was obtained from Airgas and dried using a Drierite column containing cobalt chloride and 5


4

Page 5 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Å molecular sieves. Carbon dioxide (CO2, 1.00% balance UHP N2) was obtained from Matheson. Concentrated carbon dioxide (99.9%) gas was obtained from Earlbeck Gases & Technologies. Sodium bicarbonate (99.0%, NaHCO3) was obtained from Fisher Scientific. Dimethyl methylphosphonate (DMMP, 97%), sodium carbonate (Na2CO3, 99.5%), and deuterated water (D2O, 99%) were purchased from Sigma Aldrich. UHP N2 and CO2 were dried before entering the sample chamber with two serial dry ice and acetone cold traps. Deionized (DI) water was obtained from a Thermo Fisher Scientific Barnstead GenPure ultrapure water system (18.2 MΩ cm). Thin film sample preparation. Zr(OH)4 thin films for gas phase ATR-FTIR spectroscopy were prepared by sonicating a suspension of Zr(OH)4 nanoparticles (5 mg/mL) in DI water for 4 h followed by three cycles of centrifugation at 1500 rpm and decantation. The supernatant was transferred to a clean centrifuge tube each time and any pelleted precipitate was discarded. The Zr(OH)4 nanoparticles suspension was then pipetted in air directly onto a multiple reflection ZnSe internal reflection element (IRE) and dried overnight under N2. Zr(OH)4 suspensions with D2O were prepared by drop casting the Zr(OH)4 suspension directly onto the ZnSe IRE in dry N2. In-situ ATR-FTIR spectroscopy. Gas dosing ATR experiments were performed using a Harrick Horizon accessory housed inside a Bruker Vertex 70V. During gas dosing, the sample and instrument compartments were evacuated to < 0.01 Pa to minimize infrared beam absorption from trace atmospheric gases. The sample chamber contained the Zr(OH)4 thin films sealed in a 316-stainless steel gas flow cell coated in SiO2 at ambient pressure. A gas dosing manifold containing the atmospheric component gases was attached to the flow cell via vacuum-tight bulkhead feedthrough connectors. The flow cell were purged for at least 2 h under dry UHP N2 to remove any loosely bound surface contaminants and moisture from the Zr(OH)4 surface. Interferogram phase correction was conducted using the Mertz algorithm with Happ-Genzel apodization function and zero-filling factor of 2. Unless


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 32

otherwise noted, all spectra reported are difference spectra, which were generated by subtracting the initial scan before the gas mixture reached the sample chamber from the obtained scans. No additional ATR corrections were performed. Deconvoluted spectra were fit using a Voigt line shape and OriginLab OriginPro 2017 software. Peak locations were determined using a second derivative approach. Baselines were fit using a B-spline curve. An iterative procedure to minimize the reduced Chi-square value to below 1×10-6 was used. The reduced Chi-square was obtained by dividing the residual sum of squares by the degrees of freedom. Resulting fits have an R square value greater than 0.99. Time-resolved difference spectra are colored from blue (initial) to red (final) to highlight spectral changes over time. For H2O and D2O vapor dosing, 4 scans were averaged for each spectrum and were collected at 3 s intervals for 5 min. DMMP vapor dosing was conducted by flowing UHP N2 (0.2 sccm) through a fritted gas saturation bubbler filled with DMMP. At this flow rate, the estimated concentration of DMMP reaching the sample chamber is 7,800 ppm assuming complete saturation of the bubbler headspace with DMMP and dilution to a total flow rate of 25 sccm. The bubbler was sparged overnight to purge the headspace and minimize other dissolved atmospheric gases in the DMMP. DMMP was then cointroduced with the atmospheric contaminants onto the Zr(OH)4. All exhaust gases were then fed into a bleach bubbler to decompose any remaining DMMP. The concentrations of CO2, H2O, and DMMP were intentionally reduced compared to the GCMS experiments to minimize the accumulation of multilayers of DMMP on the Zr(OH)4 surface at higher flowrates (Supporting Information, Figure S1) and prevent overtone vibrational modes from convoluting the spectra. The lower concentrations of CO2 and H2O also reduced or eliminated gas phase IR contributions, providing cleaner spectra for adsorbate identification. All samples were pre-dosed with different environmental components (N2, CO2, H2O, or air (~50% RH, ~400 ppm CO2) for 1 h to reach steady state before collecting a background spectrum and co-dosing with DMMP. The total flow rate is


6

Page 7 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


maintained throughout all experiments by offsetting flow changes with dry UHP N2. All changes to the spectra can therefore be attributed directly to DMMP interaction with pre-dosed Zr(OH)4 surfaces. Liquid phase ATR spectra were collected by pipetting ~1 mL of Zr(OH)4 suspension in H2O or D2O onto a clean ZnSe IRE in a N2 purged sample chamber. Changes in the suspension during drying were monitored by collecting spectra at 5 min intervals with 16 scans averaged per spectrum for about 4 h. ATR spectra of the bulk powders were collected using a Thermo Fisher Scientific Nicolet iS50 FT-IR spectrometer with a deuterated triglycine sulfate detector with an XT KBr beam splitter. Interferogram phase correction and scan parameters were the same as described for the gas dosed samples above, but 128 scans were averaged for each collected spectrum and background. Sample powders were placed directly onto a diamond IRE and compressed with a screw anvil. Gas chromatograph mass spectrometry (GC-MS). Zr(OH)4 (~200 mg) was placed into 2 mL GC vials in air and sealed with a Teflon coated rubber septum and screw cap. The 200 mg of Zr(OH)4 is in stoichiometric excess for DMMP hydrolysis to simulate pseudo first-order reaction kinetics. The GC vials were purged by flowing either CO2 (99.9%), UHP N2 (both dried using a dry ice/acetone cold trap), or H2O saturated UHP N2 through a non-coring needle into sealed GC vials that were vented using another non-coring needle inserted into the septum for 4 h. Once purged, concentrated DMMP (100 µL) was added by syringe through the septum immediately before sampling. The DMMP rapidly adsorbed to the Zr(OH)4 nanoparticles. Headspace samples were obtained every 30 min for 3 h on an Agilent 7890A gas chromatograph equipped with a Restek RT-Q-Bond column with helium carrier gas (6 psig) connected to an Agilent 5975C mass selective detector operating in selective ion monitoring mode with electron ionization. An Agilent 7693A autoinjector with a splitless injection equipped with a 10 µL gas tight syringe injected 5 µL of headspace from the vials. An initial temperature of 40 °C was held for 5 min then ramped to 130 °C at a rate of 7 °C/min and held for one minute with a 2 min post-run hold at


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 32

275 °C. The injection port, MS quadrupole, and source temperatures were held at 250, 150, and 230 °C, respectively. The detector monitored ions 44 and 12 m/z from 0 to 6 min, ions 17 and 18 m/z from 6 to 12.5 min, and ions 29, 30, and 31 m/z for the remainder of the run. Data were processed on Agilent Chemstation Software. Thermal gravimetric analysis with an inline mass spectrometer. A TA Instruments Discovery thermogravimetric analyzer equipped with a Discovery mass spectrometer (TGA-MS) was used to perform thermogravimetric analysis. A heating ramp of 10 °C/min from room temperature to 700 °C under N2 was employed. The MS used select ion mode to monitor ions 18 (H2O) and 44 (CO2) m/z. TGA data were processed utilizing TA Instruments Trios software. Atmospheric component dosing. All gas lines were purged prior to dosing the Zr(OH)4 thin films. All gas dosing was administered using a gas flow manifold equipped with mass flow controllers. A total flow rate of 25 sccm was achieved by balancing the atmospheric component flow streams with UHP N2. H2O and D2O dosing were accomplished by flowing UHP N2 through a gas dispersion tube submerged in either a sealed H2O, D2O, or DMMP bubbler for at least 2 h to saturate the headspace with vapor. All of the gases were sent to a mixing manifold immediately before the sample compartment to achieve a homogeneous mixture before interacting with the Zr(OH)4 samples. The RH was adjusted by diluting the saturated H2O/D2O vapor with dried UHP N2. Reported RH was calculated by the ratio of humidified gas flow rate to the total flow rate. Characterization. Powder X-ray diffraction (XRD) was performed using a Rigaku SmartLab diffractometer with a Cu Kα X-ray source set to 40 kV and 44 mA in Bragg-Brentano geometry equipped with a D/teX Ultra 250 silicon strip detector. The obtained XRD pattern was then baseline corrected. Raman spectra were obtained using a Renishaw inVia confocal Raman microscope equipped


8

Page 9 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


with a CCD detector, a 2400 lines/mm grating and 20X objective lens. The excitation source was a 785 nm diode laser. One hundred scans were averaged for each spectrum shown. The Raman spectra were obtained from dry powders placed on a silicon wafer substrate. RESULTS AND DISCUSSION Zr(OH)4 nanoparticle characterization. Zr(OH)4 is reported to have a highly disordered (amorphous) structure, resulting in broad peaks in X-ray diffraction (XRD).26 The obtained XRD pattern shows 3 broad peaks from about 20-40° and 45-65° (Figure 1A), which reasonably match previous work by Kamimura and Endo.17 The 3 sharper features near 30.1, 50.1, and 59.6° 2θ closely match the (101), (112), and (211) planes of a tetragonal ZrO2 impurity phase, respectively (JCPDF #: 42-1164). Two shoulders are also visible on each side of the tetragonal ZrO2 (101) plane along with another diffraction plane near 55.4° 2θ, coinciding with monoclinic ZrO2 (JCPDF #: 37-1484).11, 17 The broadness of these peaks indicates an amorphous crystal structure with short range crystallographic ordering.

Figure 1. Characterization of as-received Zr(OH)4 nanoparticles by (A) powder XRD, (B) ATR-FTIR spectroscopy, and (C) Raman spectroscopy. The Zr(OH)4 nanoparticles were investigated for trace and amorphous impurities and native adsorbates using ATR (Figure 1B) and Raman spectroscopy (Figure 1C). Interfacial carbonate


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 32

complexes were found on the surface, indicated by both asymmetric and symmetric ν3(OCO) modes at 1561 and 1345 cm-1, respectively.27 Additionally, a significant amount of co-adsorbed water on the Zr(OH)4 nanoparticles was identified by the δ(HOH) mode at 1634 cm-1. In general, the ν3 modes of carbonate are doubly degenerate as a free anion, but splits upon adsorption to a surface.28 Thus, the spectral separation (∆νOCO) between the split ν3 mode provides information about the adsorption strength and geometry of the carbonate, which is commonly used for identifying the bonding geometry and speciation of surface carbonates.28 Bidentate, monodentate, and polydentate carbonates have a reported ∆νOCO of about 300, 100, and 50 cm-1, respectively.29 However, as reported by Baltrusaitis, et al., interfacial carbonate complexes on Fe2O3 and γ-Al2O3 in humid environments can have a ∆νOCO near 200 cm-1.30 The ν3 carbonate modes observed on the Zr(OH)4 nanoparticles have a ∆νOCO of 216 cm-1, suggesting the observed carbonate vibrational modes are similar to interfacial carbonate complexes reported by Baltrusaitis, et. al. The detection of surface water in the ATR specta (δ(HOH) mode) further supports these observations, which is a prerequisite for solvating interfacial carbonates. Additionally, Baltrusaitis, et al. concluded that increasing the coordination of water on adsorbed carbonates (e.g. higher humidity) resulted in a reduction of the ∆νOCO of both ν3(OCO) vibrational modes. This effect was verified on Zr(OH)4 nanoparticles by dosing with D2O vapor and monitoring the spectral shift of the ∆νOCO modes (Supporting Information, Figure S2). To complement the ATR data, Raman spectra of the Zr(OH)4 nanoparticles was obtained (Figure 1C) and additional carbonate related vibrational bands were identified. Two pairs of bands were observed on the Zr(OH)4 nanoparticles at 1527/1547 cm-1 and 1345/1362 cm-1, which were assigned to the ν3(OCO)as and ν3(OCO)s stretching modes of interfacial carbonate complexes on slightly different local environments, respectively.29, 31-32 A small Raman band at 1048 cm-1 probably originates from the


10

Page 11 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


totally symmetric ν1(CO3) mode of solvated free carbonate (CO32-).33 These observations suggest at least two different types of surface carbonates occur spontaneously on Zr(OH)4 nanoparticles in air.34 Interfacial (bi)carbonate interaction. Surface carbonate assignments on Zr(OH)4 were determined by drop casting 0.1 M Na2CO3 and NaHCO3 solutions in D2O on both a ZnSe IRE and a thin film of Zr(OH)4 nanoparticles under a N2 environment (Figure 2). The features in the ATR spectra of 0.1 M NaHCO3 on a ZnSe IRE closely match the vibrational modes assigned by Su et al. for solvated NaHCO3: 1661 and 1618 cm-1 ν(OCO)as, 1362 and 1308 cm-1 ν(OCO)s, 1013 cm-1 ν(C-OH), and 843 δ(CO3)oop.35 The formation of carbonic acid intermediates was suggested in the literature,30, 36-37 but the vibrational mode corresponding to ν(OCO) of H2CO3 at 1730 cm-1 was not observed for either NaHCO3 or Na2CO3 solutions. No free carbonate anion species (CO32-) were detectable in the 0.1 M HCO3- solution. The 0.1 M Na2CO3 solution on the ZnSe IRE consisted mostly of the free carbonate anions, but the small vibrational band near 1661 cm-1 suggested a small amount of bicarbonate in the solution. This result agrees with the ~10X excess of carbonate to bicarbonate species predicted by the HendersonHasselbalch equation for a 0.1 M Na2CO3 solution at a pH of 11.3. Three vibrational bands corresponding to trigonal planar CO32- anions were observed at 1391 ν(OCO), 1067 ν(CO)s, and 883 cm-1 δ(CO3).35 The spectra of both 0.1 M solutions of Na2CO3 and NaHCO3 closely matched the FTIR spectra reported for bicarbonate and carbonate anions in water, respectively, suggesting the ZnSe IRE did not modify these solvated species.30 However, drop casting 0.1 M Na2CO3 and NaHCO3 solutions onto the Zr(OH)4 nanoparticles resulted in a shift of the ν3 modes to different energies for both the Na2CO3 (1560 and 1362 cm-1) and NaHCO3 (1549 and 1369 cm-1). Both the carbonate and bicarbonate anions adsorbed as similar solvated carbonate complexes at the Zr(OH)4 interface, regardless of which anion is predominant (HCO3- or CO32-) in solution. Such findings corroborate previous reports that NaHCO3 solutions form adsorbed carbonate complexes upon interacting with metal (hydr)oxide


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 32

surfaces.27, 35, 38 It is important to note that these spectroscopic observations are limited primarily to the interfacial region of the Zr(OH)4 due to the short penetration depth of the IR beam through the ZnSe interface (~1 µm).

Figure 2. ATR spectra comparing 0.1 M Na2CO3 and NaHCO3 in D2O drop cast onto ZnSe and Zr(OH)4 under a N2 atmosphere. Contributions from D2O and the ZnSe IRE were subtracted from the spectra. A possible deprotonation mechanism for a bicarbonate solution on metal oxides and hydroxides involves the interaction of surface hydroxyls with HCO3- as described in the equation below: M-OH + HCO3- M-O2CO- + H2O

[1]

where M is a transition metal (zirconium in this instance) forming the adsorbed carbonate complex. A similar reaction pathway was proposed and spectroscopically observed for both goethite particles39 and Fe2O3 nanopowders;30 however, in the case of the Zr(OH)4, a decrease of free zirconium hydroxyl species was observed when adding the 0.1 M NaHCO3 solution to the Zr(OH)4 thin film (Supporting Information, Figure S3). If the above proposed mechanism was solely responsible for the deprotonation of bicarbonate at the Zr(OH)4 interface, a larger decrease of free hydroxyls should have been observed.


12

Page 13 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Therefore, other mechanisms are likely participating to stabilize the adsorbed interfacial carbonate complexes on Zr(OH)4. Another possible deprotonation mechanism suggested in the literature involves the strong interaction of highly coordinated networked water with hydroxyl groups of metal oxides to enhance a surface deprotonation reaction in the adsorbed water layer instead of surface hydroxyls.27 Ab initio molecular dynamics calculations also suggested the possibility of protonic conduction between a highly coordinated H2O multilayer and the ZrO2 surface, thus providing a deprotonation pathway that does not require direct access to the Zr(OH)4 surface.40 This process has not been explored experimentally, but remains a possible deprotonation mechanism for this system. Stability of adsorbed native (bi)carbonates on Zr(OH)4. To determine the stability of adsorbates on Zr(OH)4 nanoparticles, TGA-MS was employed under an inert N2 atmosphere. Zr(OH)4 exhibited an 11% mass loss upon heating to 700 °C (Figure 3). Three distinct stages of maximum mass loss occurred at approximately 76, 153 and 409 °C. These three stages have been identified in prior literature as the loss of uncoordinated water, loss of coordinated water, and loss of water resulting from condensation of hydroxyls, respectively.13 Corresponding peaks with an 18 m/z ratio confirmed that the mass losses at 76 and 153 °C were due to water loss. However, in the third stage, only a very minor peak maximum in the H2O signal (18 m/z) occurred, which corresponded with the onset of increased CO2 signal (44 m/z) at approximately 320 °C. The CO2 signal intensity continued to increase with increasing temperature. Decomposition of bidentate carbonates was previously reported near this temperature; therefore, the observed mass loss at stage III likely combines H2O evolution from hydroxyl condensation (terminal and bridging)13 and CO2 evolution from carbonaceous species decomposition.41 Since bicarbonates typically decompose at low temperatures (200 °C)41 and desorb under vacuum from metal oxide


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 32

surfaces,27, 42 these data further support the formation of the more stable bidentate carbonates rather than bicarbonate on the surface of Zr(OH)4 under typical ambient conditions.

Figure 3. Thermal weight loss, derivative weight loss, and evolved gas from as received Zr(OH)4 determined by TGA-MS. Trace colors correspond to the color of the representative axis. It is common practice to remove or minimize adsorbates on metal (hydr)oxides by heating the sample under vacuum; however, such treatment of Zr(OH)4 significantly modifies its native surface, for example by introducing additional defect states (e.g. oxygen vacancies)43 that may alter the chemical reactivity. Moreover, formation of nanocrystalline ZrO2 domains was observed in transmission electron microscopy after heating Zr(OH)4 nanoparticles to 250 °C.44 Therefore, changes to native adsorbates were probed using in-situ ATR-FTIR under a gentle N2 purge. A suspension of Zr(OH)4 nanoparticles in DI water was drop cast onto a ZnSe IRE to form a thin film that was purged under dry N2 for 10 min at 30 °C while collecting ATR spectra (Figure 4) to monitor surface changes due to displacement of environmental CO2 and H2O with dry N2. Significant decreases in the ν(OH) stretching (~3400 cm-1), δ(HOH) bending (1636 cm-1), and broad L(HOH) libration modes (

Environmental Effects on Zirconium Hydroxide Nanoparticles and

Recommend Documents