Correlating the Network Topology of Oxide ... - ACS Publications

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Article

Correlating the Network Topology of Oxide Glasses with their Chemical Durability Nerea Mascaraque, Mathieu Bauchy, and Morten M. Smedskjaer J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.6b11371 • Publication Date (Web): 16 Jan 2017 Downloaded from http://pubs.acs.org on January 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.

The Journal of Physical Chemistry B 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 24

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

The Journal of Physical Chemistry

Correlating the Network Topology of Oxide Glasses with their Chemical Durability Nerea Mascaraque1, Mathieu Bauchy2, Morten M. Smedskjaer1,* 1

Department of Chemistry and Bioscience, Aalborg University, Fredrik Bajers Vej 7H, Aalborg 9220,

Denmark 2

Department of Civil and Environmental Engineering, University of California, Los Angeles, California

90095, USA * Corresponding author. E-mail: [email protected]

ABSTRACT: Glasses gradually dissolve and corrode when they are exposed to aqueous solutions and for many applications it is necessary to understand and predict the kinetics of the glass dissolution. Despite the recent progress in understanding the impact of chemical composition on the dissolution rate, a detailed understanding of the structural and topological origin of chemical durability in solutions of different pH is still largely lacking. Such knowledge would enable the tailoring of glass dissolution kinetics as a function of chemical composition. In a recent study focusing on silicate minerals and glasses, a direct relation was demonstrated between the dissolution rate at high pH and the number of chemical topological constraints per atom (nc) acting within the molecular network.Pignatelli, I.; Kumar, A.; Bauchy, M.; Sant, G. Langmuir 2016, 32, 4434-4439 Here, we extend this work by studying the bulk dissolution rate (Dr) of a wider range of oxide glasses in various acidic, neutral, and basic solutions. The glass compositions have been selected to obtain a wide range of chemistries and values of nc, from flexible phosphate glasses to stressed-rigid aluminosilicate glasses. We show that, in flexible glasses, the internal modes of deformation facilitate the hydration of the network, whereas, in stressed-rigid glasses, the high number of constraints largely inhibits hydration in basic, neutral and acidic solutions. Our study of chemical dissolution also allows establishing the kinetic mechanisms, which is controlled through an effective activation energy and depend on pH and glass topology. The energy barrier that needs to be overcome to break a unit atomic constraint is approximately constant for pH > 2, but then decreases at lower pH, indicating a change in dissolution mechanism from hydrolysis to ion exchange at low pH. Thus, with this research and existing topological models, the atomistic design of new oxide glasses with a specific chemical durability for a determined pH could become possible.

1 ACS Paragon Plus Environment


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 24

I. INTRODUCTION Oxide glasses are generally perceived as having a chemical durability superior to most other materials. Indeed medieval window glasses and natural glasses (e.g., obsidians and Libyan Desert glass) have survived for thousands or even millions of years1. However, oxide glasses gradually dissolve and corrode when they are exposed to an aqueous solution, which limits the range of certain applications, such as for fast ion conductors2, active lasers and amplifiers materials3, sealants for solid oxide fuel cells4, matrices for vitrification of nuclear waste5, and bioactive glasses6. These applications require accurate knowledge of dissolutions rates as a function of composition, microstructure, solution pH, and temperature,7-11 and it is therefore important to understand and predict the kinetics of oxide glass dissolution in aqueous solutions. In phosphate and silicate glasses12,13, the general reactions between these glasses and water occur in two potential steps which depend on pH and composition: i) diffusion process with ion exchange between M(+, 2+) (e.g., Na+, Ca2+) close to non-bridging oxygens (NBOs) and protonated species, and ii) hydrolysis process through the dissolution of the phosphate or silicate chains as a consequence of the breaking of P-O-P or Si-O-Si bonds into hydroxyl groups. The presence of other network-forming species (e.g., B2O3 and Al2O3) in phosphate and silicate glasses induces structural changes of the glass network, which in turn directly affects the chemical durability. In a review of the different dissolution models for aluminosilicate glasses,14 the relative contributions of each of these mechanisms have been found to be related to the content of different linkages (v: number of bridging oxygens), as, for instance, the hydrolysis of Al-O-Si bonds becomes more energetically favorable as the number of Al atoms per Si tetrahedron increases. In borosilicate glasses15,16, the introduction of boron oxide promotes the formation of BO4- groups, which are charge compensated by modifier cations, M+,

2+

. As such, the fraction of M+,

2+

cations as well of the

composition of the rest of the borosilicate network strongly affect the structure of the glass, which, in turn, can alter the dissolution rate. Moreover, the initial water content of the glasses17 and/or the presence of water in the environment can also affect the physical and chemical behavior of glassy networks. In sodium and silver phosphate glasses,18,19 the high affinity of these glasses for the ambient water molecules significantly changes their physical properties, such as the non-reversing enthalpy of relaxation the glass transition. This can be explained through the formation of –OH groups due to the hydrolysis reactions.


Page 3 of 24

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


Most of earlier dissolution studies have focused on the effect of the degree of polymerization of the network on the resistance of the glass to water attack12-14. However, additional structural features may influence chemical durability. For example, for compositions with different network former and intermediate oxides (e.g., B2O3 and Al2O3), and also different types of modifier ions (e.g., Na+ and Ca2+), not only the number of non-bridging oxygens (NBOs) is important. As an alternative approach to predict the chemical durability of oxide glasses, we therefore consider the Phillips-Thorpe topological constraint theory, as it captures more fine details of the atomic structure than the degree of depolymerization alone. Phillips and Thorpe20-22 established that the glass-forming ability of quenched liquids is determined by a balance between the number of atomic degrees of freedom and the number of interatomic chemical constraints. For a threedimensional system, each atom has three translational degrees of freedom, which are removed through the presence of rigid bond constraints, i.e., two-body bond stretching (BS) and three-body bond bending (BB) constraints. Following the Maxwell´s stability criterion23, an atomic network is classified as flexible if nc (number of topology constrains per atom) < 3, stressed-rigid for nc > 3, or isostatic for nc = 3. Smedskjaer and Bauchy have shown that, for silicate glasses, the resistance to sub-critical crack growth increases with nc, which was ascribed to a greater resistance of the network to hydrolysis reactions (stress corrosion)16. Moreover, in a recent study on the dissolution of silicate minerals and glasses, Pignatelli et al.24 demonstrated a direct relation between the dissolution rate at high pH and the number of chemical topological constraints (nc) acting between the atoms of the silicate network. In this paper, we extend the work of Pignatelli et al.24 by studying the bulk dissolution rate (Dr) and leaching rate of modifier cations of a wider range of oxide glasses in various acidic, neutral, and basic solutions. The glass compositions have been selected to obtain a wide range of chemistries and values of nc, comprising phosphate, silicophosphate, borophosphate, silicate, borosilicate, aluminosilicate, and aluminoborosilicate systems. In each case, Dr is determined by immersing the bulk samples in aqueous solutions of pH = 2, 4, 7, 10, and 14 for different times. The leaching of calcium and sodium modifier ions is determined through atomic absorption spectroscopy (AAS) measurements to study the dependence of the rate of diffusion reaction on pH and glass chemistry. In this paper, we establish a direct relationship between dissolution kinetics, glass composition, and network topology in aqueous solutions of different pH values.



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 24

II. THEORY Seven oxide glass compositions were selected to obtain a wide range of nc values, based on earlier studies2528

. These glasses comprise different types of network forming and intermediate species, which allows us to

assess the effect of various structural features on chemical durability. The selected compositions are 36Na2O.64P2O5,

30Na2O.20SiO2.50P2O5,

50CaO.5B2O3.45P2O5,

15Na2O.10CaO.75SiO2,

15Na2O.10CaO.15B2O3.60SiO2, 15Na2O.10CaO.15Al2O3.60SiO2, and 15Na2O.17Al2O3.5B2O3.63SiO2 (in mol.%), and are referred to as P, SiP, BP, Si, BSi, AlSi, and AlBSi, respectively (Table 1). To avoid relying on unproven guesses, the enumeration of the topological constraints critically relies on the accurate knowledge of the concentration and coordination number (ri) of each species that is part of the atomic network. Indeed, the number of BS and BB constraints are usually given by ri/2 (note that each BS constraint is shared by two atoms) and 2ri – 3 (the number of independent angles needed to define a polyhedron), respectively. In the case of the considered oxide glasses, the radial and angular constraints associated to network-forming atoms usually remain intact at room temperature25-29. In sodium phosphosilicate and phosphate glasses25, two and one radial BS constraints (Si-O and P-O) are created by each bridging oxygen (BO) and NBO, respectively. Regarding angular constraints, each BO (Si-O-Si, Si-O-P, and P-O-P) feature one BB constraint. Note that, in contrast, the angular constraints associated to NBO atoms are thermally broken at room temperature29. In addition, nine BB constraints are associated to Si6 units, five to Si4, Si3, Si2, Si1, Si0, P2, P1, and P0 units, and three to P3 units. Finally, network modifying cations also contribute to the rigidity of the network. Namely, one additional BS constraint is associated to each Na+-NBO bond. In sodium and calcium borophosphate glasses26, two and one radial BS constraints (B-O and P-O) are created by each BO and NBO, respectively, and each BO features an additional BB constraint (B-O-B, B-OP, and P-O-P), as in the case of phosphosilicate glasses. Moreover, three angular BB constraints (O-B-O and O-P-O) are found around B3, B2, B1, B0, and P3 units, and five in the case of B4, P4, P2, P1 and P0 species. Finally, one BS constraints is associated to each Ca2+-NBO bond. In sodium and calcium silicate glasses comprising Al2O3 and/or B2O3,27-28 each oxygen shows two radial BS constraints (Si-O, Al-O, B-O, and MNBO-O) and one BB constraint (Si-O-Si, Al-O-Al, B-O-B, SiO-B, Si-O-Al, Al-O-B, Si-O-MNBO, Al-O-MNBO and B-O-MNBO), where MNBO refers to NBO-forming Na or


Page 5 of 24

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


Ca species. In addition, angular constraints comprise five BB constraints per Si (O-Si-O), seven ΒΒ constraints per Al5 (O-Al-O), five BB constraints per Al4, five BB constraints per B4 (O-B-O), and three BB constraints per B3. Finally, each MNBO involves the formation of two clustering constraints. We note that in an earlier study, a different topological model has been suggested for the soda lime silicate “window” glass, which found this composition to be isostatic.30 However, for consistency, we use the same constraint enumeration protocol, which has been validated in the earlier studies.25-28 Based on earlier nuclear magnetic resonance (NMR) and x-ray photoelectron spectroscopy (XPS) experiments for each glass system,25-27,31 the fractions of the different types of silicon, phosphorus, boron, and aluminum species have been quantified. This knowledge, combined with the topological models presented above, allows us to calculate the values of nc for each glass. Note that, for all glasses but AlSi, the details of the constraints enumeration is presented in Refs.25-28 In the case of the AlSi glass, we have calculated nc following the topological model of boroaluminosilicate glasses28 by taking into account that all Na+ are used for charge balancing Al4. That is, Ca2+ only acts to create NBOs in the structure, thereby resulting in two modifier constraints per Ca2+-NBO. Table 1 summarizes the glass compositions and calculated values of nc, which covers a wide range of values, from the flexible phosphate glass (nc = 2.76) to the stressed-rigid boroaluminosilicate glass (nc = 3.63).

III. EXPERIMENTAL SECTION The seven glasses were prepared using the melt-quenching method by mixing batches of the following reagent grade materials: Na2CO3 (99.999% Suprapur®, Merck), CaCO3 (Reag. Ph Eur, Merck), NH4H2PO4 (≥98% ACS Reagent, Sigma-Aldrich), υ-B2O3 (B2O3 previously melted at 850°C using H3BO3 as raw material), SiO2 (Sigma-Aldrich, purum p.a.), and Al2O3 (Sigma-Aldrich, ≥99%). In the case of the P2O5based glasses (P, SiP, and BP), the batches with their stoichiometric amounts were first calcined in porcelain crucibles up to 400 °C in an electric furnace overnight, and then melted for 2 h at 750, 950, and 1100 °C for P, SiP, and BP glasses, respectively. The melts were cast onto brass plates and the transparent glasses were annealed for 30 minutes below their respective glass transition temperature (Tg), which was determined using differential scanning calorimetry. For BSi and Si glasses, the mixed batches were melted and homogenized in a covered Pt90Rh10 crucible at 1400-1450 °C for 2 h in an inductively heated furnace. To prepare AlSi and 5 ACS Paragon Plus Environment


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

AlBSi glasses, the batch materials were first thoroughly mixed for 60 min using a ball milling, and then melted in covered Pt crucibles at 1650 °C for 6 h in air. In order to improve the chemical homogeneity of these high-viscosity melts, they were first quenched in water and the resulting glass shards were crushed and re-melted for another 6 h at 1650 °C. Finally, all the SiO2-based glass melts were poured onto a stainless steel plate in air and then annealed at their respective Tg. To assess the chemical durability of the glasses, 10 samples of each oxide glass were cut to rectangular shape, and their faces were polished using 4000-grit SiC paper at the final step. Prior to the measurements, the samples were cleaned with distilled water under ultrasonication for 5 min and then with acetone under ultrasonication for 5 min, and finally dried in an oven for 1 h. Afterwards, the weight and surface area (length x width x thickness) were measured. Aqueous solutions (50 mL) of 0.02 mM HCl, 2 mM HCl, distilled water, 0.05 mM KOH, and 0.5 M KOH were used to obtain pH values of 2, 4, 7, 10, and 14, respectively. For each experiment, the solid-to-liquid ratio was selected to avoid any saturation effect. The bulk dissolution rate of each glass was determined by measuring the weight loss after immersion of the bulk sample in basic (pH = 10 and 14), acidic (pH = 2 and 4), and neutral (pH = 7) aqueous solutions at room temperature as a function of time. The experiment was repeated twice for each glass composition in each solution. The pH of the media was measured as a function of dissolution time using a pH meter (M220X9541) and maintained constant throughout the experiments by adding basic or acid solutions. Note that the duration of the dissolution test was adjusted for each glass (from 6 h to nearly two months), to achieve substantial values of weight loss. In addition to studying the bulk dissolution rates of the samples, the leaching of calcium and sodium ions into the pH 2 and 7 solutions was also investigated after the end of bulk dissolution testing, i.e., after the longest immersion time. The concentrations of Na+ and Ca2+ (in ppm) were determined using atomic absorption spectrometry (AAnalyst 100, Perkin Elmer). Results are given in Table S1 in the Supporting Information.

IV. RESULTS AND DISCUSSION We test the chemical durability of the seven glass compositions in five different solutions with pH 2, 4, 7, 10 and 14. Due to the large number of durability experiments (70), we here only show the weight loss as a function of time for the SiP glass (Figure 1). The results for the six other glasses are shown in Figures S1-S6 6 ACS Paragon Plus Environment

Page 6 of 24

Page 7 of 24

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


in the Supporting Information. For each pH value, two samples are analyzed in order to ensure the reproducibility of the experiments. To permit consistent comparisons among glasses, all values of weight loss are normalized by the area of the surface initially exposed. Overall, despite some initial non-linearity at short times, we observe that a linear increase of the weight loss over time, for each sample and each pH. In each case, the absence of any plateau demonstrates that each solution remains far from saturation over the course of the dissolution test, so that only the structure of the glass is controlling the kinetics of the dissolution. We note that the Si glass consistently features the largest deviation from linearity, at all pH values. As mentioned below, this is likely to arise from the formation of a protective hydration layer at the surface of the glass, which results in a decrease of the dissolution rate over time, which also occurred in the BSi glass at high pH. In contrast, for all other glasses, the linear behavior of the weight lost suggests the absence of any protective layer, so that the structure of the pristine glasses should control the kinetics of their dissolution. Dissolution can occur through various processes, comprising surface hydrolysis, hydration, or ion exchange, which can occur individually or in combination.12 As per the Fick’s law, processes involving the diffusion of species are expected to show a square root dependence on time. As such, the linear dependence over time of the weight loss observed herein suggests that sample experience a linear surface retreat over time, so that, for all glasses and pH, the kinetics of the dissolution is rate-limited by the surface hydrolysis step at long time. In the following, we describe the durability data separately for the flexible and stressesrigid glasses. The chemical durability data of all the samples are compiled in Figure S7 in the Supporting information.

A. Dissolution behavior of flexible and isostatic networks (nc ≤ 3) The glasses referred to as P and SiP are flexible and BP is essentially isostatic and they are thus less rigid compared to the silica-rich glasses with higher nc values (see Table 1). Figure 2 shows the pH dependence of the logarithmic dissolution rate Dr for these three flexible and isostatic glasses. Dr is calculated from the slope of linear fits to the weight loss curves (Figs. 1 and S1-S6). As demonstrated in Fig. 2, there is a minimum in Dr around the neutral solution (pH=7) for the three glasses, i.e., a clear increase of dissolution rate (or decrease of chemical resistance) when increasing or decreasing pH compared to neutral conditions.



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

The 31P NMR spectrum of the studied P glass composition reveals the presence of P3 and P2 species in equal amounts.25 The P3 and P2 units represent phosphorus environments with 3/2BO+1NBO and 1BO+2NBO, respectively, i.e., the estimated fractions of BO and NBO are 45% and 55%, respectively. From the deconvolution of 29Si and 31P NMR spectra of the SiP glass composition,25 it has been calculated that there are 8% Si6, 8% Si4, 26% P2, and 58% P3, with 64% BO and 36% NBO (taking into account that Si6 and Si4 units present 6 and 4 BOs, respectively). In both P and SiP glasses, the modifier cation is Na+ which only acts to charge balance NBOs on P3 and P2 species. Thus, the main factor controlling the dissolution behavior for these glasses is presumably the degree of phosphate network connectivity. A higher content of NBOs, i.e., a higher degree of network depolymerization, produces a decreased resistance to water attack (higher log Dr), as observed by comparing Figures 2a and 2b. In the BP glass, the addition of B2O3 gives rise to an increase of NBO and the formation of only B4, as observed in its

11

B NMR spectrum.26 However, the

introduction of boron oxide leads to an increase of the degree of cross-linking in the glass network through formation of P-O-B bonds, presenting higher resistance to water attack in the more rigid network glasses (Figure 2c). The dissolution mechanism of these P2O5-containing glasses could be explained through the reactions of ion-exchange (1) and hydrolysis (2), which are strongly coupled.12 However, each reaction depends predominantly on pH and the glass topology. P-O--M+ + H2O → P-OH + M+ + OH-

(1a)

P-(O-)2-M2+ + 2H2O → P-(OH)2 + M2+ + 2OH-

(1b)

P-O-P + H2O → 2 (P-OH)

(2a)

P-O-B + H2O → P-OH + B-OH

(2b)

In acidic solutions (pH = 2 and 4), reaction (1) is favored as the ion-exchange step is a process of protonation of phosphate chains: P-O--M+ + H+ → P-OH + M+

(3a)

P-(O-)2-M2+ + 2H+ → P-(OH)2 + M2+

(3b)

The faster protonation reaction at pH = 2 is verified through the results of the leaching study (Table S1 in Supporting Information). A higher concentration of sodium or calcium is measured in the pH = 2 solution compared to that in pH = 7, implying that the rate of the diffusion process (3) is faster for lower pH (acid media). 8 ACS Paragon Plus Environment

Page 8 of 24

Page 9 of 24

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


In basic solutions (pH = 10 and 14), reaction (2) is preferred because P-O-P and P-O-B are susceptible to nucleophilic attack of OH- (hydrolysis), which produces the following degradation: P-O-P + OH- → P-OH + P-O-

(4a)

P-O-B + OH- → P-OH + B-O-

(4b)

The reactions (3) and (4) induce a faster corrosion in acid and basic solutions, respectively, of the P, SiP and BP glasses compared to the neutral aqueous media (Figure 2). Nevertheless, the BP glass exhibits a different pH dependence of dissolution rate compared to the SiP and P glasses. That is, for the borophosphate glass, the chemical durability at pH = 14 (basic) is higher compared to that at pH = 2 (acid), while the opposite behavior is observed for P and SiP glasses. This could be because the nucleophilic attack in reaction (4) is more impeded in glasses with a more cross-linked network, such as the BP glass.

B. Dissolution behavior of stressed-rigid networks (nc > 3) The Si, BSi, AlSi, and AlBSi glasses all have nc values larger than 3 (Table 1). Figure 3 presents the pH dependence of the logarithmic dissolution rate Dr for these glasses. For the BSi glass,

29

Si and

11

B NMR

spectra have revealed the presence of 70% of the total silicon atoms are as Si3, and 76% of the total boron atoms as B4.27 Most of the NBOs are associated with SiO4 tetrahedra. In the AlSi glass, aluminum is only present as Al4 and fully compensated by Na+, i.e., Ca2+ only acts to create NBOs in silicate structure.32 Both BSi and AlSi glasses contain 15 mol.% Na2O and 10 mol.% CaO, but there is a difference in the role of these modifiers in each glass. In BSi glass, both sodium and calcium act to create NBOs in the silicate network and covert boron from B3 to B4. However, in the AlSi glass, only Ca2+ acts to create NBOs in silicate structure, i.e., the degree of depolymerization of the silicate network is lower. 29Si, 27Al and 11B NMR spectra of the AlBSi glass have shown the presence of Al4, B3, and Si4 species, with all Na+ cations used for charge balancing Al4.31 Based on these structural differences, it can be inferred that the polymerization degree of the silicate network increases from BSi to AlSi to AlBSi. Thus, a lower value of log Dr, i.e., better chemical durability, is expected for AlBSi glass, as confirmed in Figure 3. The results in Figure 3 also show that the dissolution rate decreases with increasing pH from 2 to 7 for all four glasses, whereas the pH dependence of Dr depends on the glass composition for pH>7. Moreover, for the Si glass, the values of Dr appear independent of the solution pH value (Figure 3a). As for the flexible network glasses, the dissolution mechanism could be explained through the reactions of ion-exchange (5) 9 ACS Paragon Plus Environment


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 24

and hydrolysis (6), together with the possible reaction of re-polymerization (7) that produces the formation of a white layer on the glass surface.33 The latter is a consequence of the reaction between silanol groups, which are previously formed via ion-exchange (5) and hydrolysis (6). The reaction of condensation (7) seems to be faster in silicate systems with higher concentration of NBOs associated to Si atoms, since the attack of water via ion-exchange (5) is faster in systems with a lower degree of network cross-linking, i.e., higher amount of Si-O- bonds. (Si, B or Al)-O--M+ + H2O → (Si, B or Al)-OH + M+ + OH-

(5a)

(Si, B or Al)-(O-)2-M2+ + 2H2O → (Si, B or Al)-(OH)2 + M2+ + 2OH-

(5b)

Si-O-(Si, B or Al) + H2O → 2 (Si, B or Al)-OH

(6)

2 (Si-OH) → Si-O-Si + H2O

(7)

Indeed, in the case of the Si glass, we observed initially the formation of a white layer on the surface following immersion of the samples in all pH solutions. According to the observation of Bunker8, alkali leaching is not a reversible ion-exchange reaction (5) due to the formation of Si-O-Si bonds (condensation reaction (7)), causing a complete restructuring of glass and appearance of the white layer on the surface. Thus, this layer could be the responsible for the slow dissolution rate of the Si glass in all the different pH solutions (Figure 3a). The same white layer is also observed on the surface of the BSi glass for pH > 7, which could explain why the Dr value for this glass is constant for 7 2. Indicating that dissolution through ion-exchange become predominant at pH =2, in agreement with the conventional wisdom in the field.35 In addition, the highest


Page 13 of 24

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


values of activation energy are reached at pH = 14, showing that dissolution through hydrolysis become predominant for high pH.

V. CONCLUSIONS The dissolution kinetics of seven oxide glasses with different network former and intermediate cations have been investigated through measurements of weight loss of bulk samples immersed in acid (pH = 2 and 4), neutral (pH = 7), and basic (pH = 10 and 14) solutions. A direct relationship between their chemical durability and the network topology has been found. In general, for the same pH value, the increase of the number of constraints per atom in oxide glasses improves the dissolution resistance. The activation energy that controls the kinetics of the dissolution is observed to decrease at low pH, while it is remains approximately constant at pH > 2, suggesting that a change from dissolution through hydrolysis to ion exchange occurs at low pH in agreement with many earlier studies. Based on these results, we demonstrate that the rate-limiting step of dissolution, and its activation energy, depends on the pH of the solution. Altogether, this study shows that topological constraint theory could be used to discover new oxide glasses with tailored bulk dissolution rates at a targeted pH.

ASSOCIATED CONTENT Supporting Information. Weight loss curves for the six glasses in different pH solutions; pH dependence of dissolution rate; fraction of NBOs dependence of dissolution rate; atomic constraint dependence of dissolution rate; concentrations of leached sodium and calcium ions. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author. *E-mail: [email protected] Notes. The authors declare no competing financial interest.

ACKNOWLEDGEMENTS N.M. and M.M.S. acknowledge financial support from VILLUM Fonden (Postdoctoral Block Fellowship Program). M.B acknowledges support from the National Science Foundation (Grant No. 1562066). 13 ACS Paragon Plus Environment


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

REFERENCES (1) Henderson, J. Ancient Glass: An Interdisciplinary Exploration; Cambridge University Press, University of Nottingham, 2013. (2) Pradel, A.; Ribes, M. Ionic Conductive Glasses. Mater. Sci. Eng. B 1989, 3, 45-56. (3) Ehrt, D. Fluoroaluminate Glasses for Lasers and Amplifiers. Curr. Opinion Solid State Mater. 2003, 7, 135-141. (4) Kumer, V.; Arora, A.; Pandey, O. P.; Singh, K. Studies on thermal and structural properties of glasses as sealants for solid oxide fuel cells. Int. J. Hydrog. Energy 2008, 33, 434-438. (5) Weber, W. J.; Navrotsky, A.; Stefanovsky, S.; Vance, E. R.; Vernaz, E. Materials Science of High-Level Nuclear Waste Immobilization. MRS Bull. 2009, 34, 46-53. (6) Jones, J. R. Review of bioactive glass: From Hench to hybrids. Acta. Biomater. 2013, 9, 4457-4486. (7) Wolff-Boenisch D.; Gislason, S. R.; Oelkers, E. H.; Putnis, C. V. The dissolution rates of natural glasses as a function of their composition at pH 4 and 10.6, and temperatures from 25 to 74°C. Geochim. Cosmochim. Acta 2004, 68, 4843-4858. (8) Angeli, F.; Gaillard, M.; Jollivet, P.; Charpentier, T. Influence of glass composition and alteration solution on leached silicate glass structure: A solid-state NMR investigation. Geochim. Cosmochim. Acta 2006, 70, 2577-2590. (9) Cailleteau, C.; Angeli, F.; Devreux, F.; Gin, S.; Jestin, J.; Jollivet, P.; Spalla, O. Insight into silicate-glass corrosion mechanisms. Nat. Mater. 2008, 7, 978-983. (10) Gin, S.; Beaudoux, X.; Angeli, F.; Jegou, C.; Godon, N. Effect of composition on the short-term and long-term dissolution rates of ten borosilicate glasses of increasing complexity from 3 to 30 oxides. J. Non-Cryst. Solids 2012, 358, 2559-2570. (11) Hopf, J.; Kerisit, S. N.; Angeli, F.; Charpentier, T.; Icenhower, J. P.; McGrail, B. P.; Windisch, C. F.; Burton, S. D.; Pierce, E. M. Glass–water interaction: Effect of high-valence cations on glass structure and chemical durability. Geochim. Cosmochim. Acta 2016, 181, 54-71. (12) Bunker, B. C.; Arnold, G. W.; Wilder, J. A. Phosphate glass dissolution in aqueous solutions. J. NonCryst. Solids 1984, 64, 291-316.


Page 14 of 24

Page 15 of 24

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


(13) Bunker, B. C. Molecular mechanisms for corrosion of silica and silicate glasses. J. Non-Cryst. Solids. 1994, 179, 300-308. (14) Hamilton, J. P.; Brantley, S. L.; Pantano, C. G.; Criscenti, L. J.; Kubicki, J. D. Dissolution of nepheline, jadeite and albite glasses: toward better models for aluminosilicate dissolution. Geochim. Cosmochim. Acta 2001, 65, 3683-3702. (15) Bunker, B. C.; Tallant, D. R.; Headley, T. J.; Turner, G. L.; Kirkpatrick, R. J. Structure of Leached Sodium Borosilicate Glass. Phys. Chem. Glasses 1988, 29, 106-120. (16) Smedskjaer, M. M.; Bauchy, M. Sub-critical crack growth in silicate glasses: Role of network topology. Appl. Phys. Lett. 2015, 107, 141901. (17) Bhosle, S.; Gunasekera, K.; Boolchand, P.; Micoulaut, M. Melt Homogenization and Self-Organization in Chalcogenides-Part II. Int. J. Appl. Glass Sci. 2012, 3, 205-220. (18) Novita, D.; Boolchand, P. Synthesis and structural characterization of dry AgPO3 glass by Raman scattering, infrared reflectance, and modulated differential scanning calorimetry. Phys. Rev. B 2007, 76, 184205. (19) Fabian Jr., R.; Sidebottom D. L. Dynamic light scattering in network-forming sodium ultraphosphate liquids near the glass transition Phys. Rev. B 2009, 80, 064201. (20) Phillips, J. C. Topology of covalent non-crystalline solids I: Short-range order in chalcogenide alloys. J. Non-Cryst. Solids 1979, 34, 153-181. (21) Phillips, J. C.; Thorpe, M. F. Constraint theory, vector percolation and glass formation. Solid State Commun. 1985, 53, 699-702. (22) He, H.; Thorpe, M. F. Elastic Properties of Glasses. Phys. Rev. Lett. 1985, 54, 2107-2110. (23) Maxwell, J. C. L. On the calculation of the equilibrium and stiffness of frames. Philos. Mag. Ser. 4 1864, 27, 294-299. (24) Pignatelli, I.; Kumar, A.; Bauchy, M.; Sant, G. Topological Control on Silicates’ Dissolution Kinetics. Langmuir 2016, 32, 4434-4439. (25) Hermansen, C.; Guo, X. J.; Youngman, R. E.; Mauro, J. C.; Smedskjaer, M. M.; Yue, Y. Z. Structuretopology-property correlations of sodium phosphosilicate glasses. J. Chem. Phys. 2015, 143, 064510. (26) Hermansen, C.; Youngman, R. E.; Wang, J.; Yue, Y. Z. Structural and topological aspects of borophosphate glasses and their relation to physical properties. J. Chem. Phys. 2015, 142, 184503. 15 ACS Paragon Plus Environment


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

(27) Smedskjaer, M. M., Mauro, J. C.; Youngman, R. E., Hogue, C. L., Potuzak, M., Yue, Y. Z. Topological Principles of Borosilicate Glass Chemistry. J. Phys. Chem. B 2011, 115, 12930-12946. (28) Smedskjaer, M. M. Topological model for boroaluminosilicate glass hardness. Front. Mater. 2014, 1, 23. (29) Bauchy, M.; Micoulaut, M. Atomic scale foundation of temperature-dependent bonding constraints in network glasses and liquids. J. Non-Cryst. Solids 2010, 357, 2530-2537. (30) Kerner, R.; Phillips, J. C. Quantitative principles of silicate glass chemistry. Solid State Commun. 2000, 117, 47-51. (31) Zheng, Q. J.; Youngman, R. E.; Hogue, C. L.; Mauro, J. C.; Potuzak, M.; Smedskjaer, M. M.; Yue, Y. Z. Structure of boroaluminosilicate glasses: impact of [Al2O3]/[SiO2] ratio on the structural role of sodium. Phys. Rev. B 2012, 86, 054203. (32) Amma, S.; Lanagan, M. T.; Kim, S. H.; Pantano, C. G. Ionic Conductivity in Sodium–Alkaline Earth– Aluminosilicate Glasses. J. Am. Ceram. Soc. 2016, 99, 1239-1247. (33) Dugger, D. L.; Stanton, J. H.; Irby, B. N.; McConnell, B. L.; Cummings, W. W.; Maatman, R. W. The Exchange of Twenty Metal Ions with the Weakly Acidic Silanol Group of Silica Gel. J. Phys. Chem. 1964, 68, 757-760. (34) Naumis, G. G. Energy landscape and rigidity. Phys. Rev. E 2005, 71, 026114. (35) Varshneya, A. K. Fundamentals of Inorganic Glasses; Society of Glass Technology: Sheffield, U.K., 2006.


Page 16 of 24

Page 17 of 24

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


FIGURE CAPTIONS Figure 1. Weight loss (in mg·dm-2) of the SiP glass as a function of time in different aqueous solutions: (a) pH=2, (b) pH=4, (c) pH=7, (d) pH=10, (e) pH=14. Two repeated experiments are performed for each combination of glass and solution pH. The dashed lines represent linear fits to the data. Figure 2. pH dependence of the logarithmic dissolution rate (Dr, in mg·dm-2·h-1) for the flexible and isostatic glasses: (a) P glass, (b) SiP glass, (c) BP glass. Figure 3. pH dependence of the logarithmic dissolution rate (Dr, in mg·dm-2·h-1) for the stressed-rigid glasses: (a) Si glass, (b) BSi glass, (c) AlSi glass, (d) AlBSi glass. Figure 4. Logarithmic dissolution rate (Dr, in mg·dm-2·h-1) of the studied oxide glass as a function of the number of constraints per atom (nc) in different aqueous solutions: (a) pH=2, (b) pH=4, (c) pH=7, (d) pH=10, (e) pH=14. The dashed lines are drawn as guides for the eyes. Insets: Dr on linear scale as a function of nc. Error bars are on the size of the symbols. Figure 5. Energy barrier needed to break a unit constraint (E0) as a function of solution pH. E0 is determined by fitting the data in Fig. S9 in Supporting Information to Eqs. (10) and (11).

TABLE CAPTIONS Table 1. Nominal compositions (in mol. %) and calculated number of constraints per atom (nc) of the investigated glasses.

Table 2. Values of

for all the samples in a wide range of pH (from 2 to 14).



Figure 1 (a)

(b)

6000

3000

pH=4

-2

Weight loss (mg dm )

-2


pH=2 4800 3600 2400 1200

exp. 1 exp. 2

0 0

20

40

60

80

2400 1800 1200 600

exp. 1 exp. 2

0 0

100 120 140 160

20

40

60

(d) 3000


pH=7

-2

2500 2000 1500 1000 500

exp. 1 exp. 2

0 0

20

40

60

80

100 120 140 160

pH=10

2500 2000 1500 1000 500

exp. 1 exp. 2

0 0

100 120 140 160

20

40

60

Time (h)

80

100 120 140 160

Time (h)

(e) 10000 pH=14 -2

-2

3000


(c)

80

Time (h)

Time (h)


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 18 of 24

8000 6000 4000 2000

exp. 1 exp. 2

0 0

20

40

60

80

100 120 140 160

Time (h)


Page 19 of 24

Figure 2

-1

1.8

SiP glass (nc=2.90)

-2

2.50

log Dr (Dr in mg dm h )

P glass (nc=2.76)

2.25 2.00 1.75 1.50 1.25

1.6 1.4 1.2 1.0

1.00 2

4

6

8

10

12

14

0

16

2

4

6

8

pH (-)

pH (-) (c) 1.6 -1

0

BP glass (nc=3.01)

1.2

-2

-1 -2

(b)

2.75


(a)


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


0.8 0.4 0.0 -0.4 -0.8 0

2

4

6

8

10

12

14

pH (-)


16

10

12

14

16


Figure 3 (b)

0.0

3

BSi glass (nc=3.35)

-1 -2

-0.5


Si glass (nc=3.26)

-2

-1


(a)

-1.0 -1.5 -2.0 -2.5 -3.0

2 1 0 -1 -2 -3

0

2

4

6

8

10

12

14

16

0

2

4

6

pH (-)

(d)

2

0 -1 -2 -3 2

4

6

8

10

12

14

16

14

16

2

AlBSi glass (nc=3.63)

-1

1

-2

-2

-1

AlSi glass (nc=3.59)

0

8

pH (-)


(c) log Dr (Dr in mg dm h )

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 20 of 24

10

12

14

1 0 -1 -2 -3

16

0

2

4

pH (-)

6

8

pH (-)


10

12

Page 21 of 24

Figure 4 flexible

3

(b)

stressed-rigid

-2

40 0 2.6

2.9

3.2

3.5

3.8

nc (-)

-3 2.6

2.8

3.0

3.2

3.4

3.6

40 -1

-2

30 20

-4

10 0 2.6

2.9

3.2

3.5

2.8

3.0

nc (-) flexible

2

(d)

stressed-rigid

pH=7

3.4

3.6

3.8

flexible

2

stressed-rigid

pH=10

-1 10

3.2

3.5

3.8

nc (-)

3.2

3.4

3.6

-1

20

-4

10 0 2.6

-6 2.6

3.8

2.9

3.2

3.5

3.8

nc (-)

2.8

3.0

nc (-)

3.2

nc (-)

(e)

flexible

stressed-rigid

pH=14

2

400 -1

-2

300 200

-4

100 0 2.6

-6 2.6

flexible stressed-rigid

0

-2

3.0

Dr (mg dm h )

2.8

-1

-6 2.6

2.9

30

-2

0 2.6


-4

40

-2


20

0

-2

30

-2

-1

Dr (mg dm h )

40

-2


-2

0

Dr (mg dm h )

-1 -2

3.2

nc (-)


(c)

3.8

nc (-)

-6 2.6

3.8


80

0

-2

120

-2

-1

Dr (mg dm h )

160


1

-1

stressed-rigid

pH=4

Dr (mg dm h )

-2

2

0

flexible

2

-1


pH=2

-2

-1


(a)


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


2.9

3.2

3.5

3.8

nc (-)

2.8

3.0

3.2

3.4

3.6

nc (-)


3.8

3.4

3.6

3.8


Figure 5 30

E0 (kJ/mol)

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 22 of 24

20

10

0 0

2

4

6

8

10

12

pH (-)


14

Page 23 of 24

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


Table 1 Glass ID

Composition (in mol.%)

System

nc (-)

Reference

P

36Na2O.64P2O5

Phosphate

2.76

25

SiP

30Na2O.20SiO2.50P2O5

Silicophosphate

2.90

25

BP

50CaO.5B2O3.45P2O5

Borophosphate

3.01

26

Si

15Na2O.10CaO.75SiO2

Silicate

3.26

27

BSi

15Na2O.10CaO.15B2O3.60SiO2

Borosilicate

3.35

27

AlSi

15Na2O.10CaO.15Al2O3.60SiO2

Aluminosilicate

3.59

this work

AlBSi

15Na2O.17Al2O3.5B2O3.63SiO2

Aluminoborosilicate

3.63

28

Table 2

nc (-)

Effective activation energy,

(kJ/mol)

pH=2

pH=4

pH=7

pH=10

pH=14

2.76

24.9

56.9

55.0

58.4

68.1

2.90

26.2

59.8

57.8

61.4

71.6

3.01

27.2

62.1

60.0

63.7

74.3

3.26

30.2

69.1

66.8

70.9

82.7

3.35

32.4

74.0

71.5

76.0

88.6

3.59

32.8

74.9

72.3

76.9

89.6



-2

-1

300 200 100


TABLE OF CONTENT IMAGE

Dr (mg dm h )

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

0 2.6

2.9

3.2

3.5

3.8

nc (-)


Page 24 of 24

Correlating the Network Topology of Oxide ... - ACS Publications

Recommend Documents