Direct Experimental Evidence for Abundant BO4âBO4 Motifs in

Subscriber access provided by RMIT University Library

Chemical and Dynamical Processes in Solution; Polymers, Glasses, and Soft Matter

Direct Experimental Evidence for Abundant BO#--BO# Motifs in Borosilicate Glasses From Double-Quantum ¹¹B NMR Spectroscopy Yang Yu, Baltzar Stevensson, and Mattias Edén J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.8b02907 • Publication Date (Web): 16 Oct 2018 Downloaded from http://pubs.acs.org on October 22, 2018

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 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 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.

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 15 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 Letters

Direct Experimental Evidence for Abundant BO4–BO4 Motifs in Borosilicate Glasses From Double-Quantum

11

NMR Spectroscopy Yang Yu, Baltzar Stevensson, and Mattias Ed´en∗

Physical Chemistry Division, Department of Materials and Environmental Chemistry, Stockholm University, SE-106 91 Stockholm, Sweden

∗

Corresponding author. E-mail: [email protected]

1 Environment ACS Paragon Plus

B

The Journal of Physical Chemistry Letters 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 15

Abstract By using double-quantum–single-quantum correlation

11

B nuclear magnetic resonance (NMR)

experiments and atomistic molecular dynamics (MD) simulations, we resolve the long-standing controversy of whether directly interlinked BO4 –BO4 groups exist in the technologically ubiquitous class of alkali/alkaline-earth based borosilicate (BS) glasses. Most structural models of Na2 O– B2 O3 –SiO2 glasses assume the absence of B[4] –O–B[4] linkages, whereas they have been suggested to exist in Ca-bearing BS analogs. Our results demonstrate that while B[4] –O–B[4] linkages are disfavored relative to their B[3] –O–B[3] /B[4] counterparts, they are nevertheless abundant motifs in Na2 O–B2 O3 –SiO2 glasses over a large composition space, while the B[4] –O–B[4] contents are indeed elevated in Na2 O–CaO–B2 O3 –SiO2 glasses. We discuss the compositional and structural parameters that control the degree of B[4] –O–B[4] bonding. TOC GRAPHIC


Page 3 of 15 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


Thanks to the relatively large glass-forming regions of most borosilicate-based systems and thereby tunable physical and chemical glass properties, such as hardness, refractive index, thermal expansion, and corrosion resistance, these glasses feature a multitude of applications, for instance as flat-screen displays, ion conductors, and laboratory glassware (e.g., ”Pyrex”), as well as materials for radioactive waste encapsulation and biomedical implants.1–3 Several structural theories/models are proposed for borosilicate (BS) glasses incorporating alkali/alkaline-earth metal ions,4–9 such as the herein targeted Na2 O–(CaO)–B2 O3 –SiO2 glasses. The basic building blocks and the short-range structure of BS glasses are well understood: their networks comprise planar BO3 triangles interconnected with [BO4 ]− and SiO4 tetrahedra via bridging oxygen (BO) atoms, while electropositive cations (e.g., Na+ and Ca2+ ) assume a dual structural role of balancing the negatively charged [BO4 ]− tetrahedra, and as network modifiers that depolymerize the glass network by introducing non-bridging oxygen (NBO; O− ) species.6 Yet, despite vast experimental efforts to unveil BS glass structures,4–19 many basic aspects of their medium-range organization are incompletely understood and remain debated, such as the preferential silicate/borate-group intermixing across the networks, or whether ”superstructural borate” units (known to occur in borate glasses6 ) and structural inhomogeneities are present across a (sub)nm scale;7–14 here the presence of borate and (boro)silicate domains have been suggested in ”Pyrex”13, 14 and in P-bearing BS glass networks.20, 21 A fundamental but hitherto unsettled feature of BS glasses concerns the precise structural role of the [BO4 ]− tetrahedra, whose negative charges are usually believed to preclude both B[4] –NBO bonds and B[4] –O–B[4] linkages to avoid local charge accumulations across the BS network;4–6 this is directly analogous with the well-known ”Loewenstein Al[4] avoidance” rule,22 where the absence of Al[4] –O–Al[4] motifs imposes strictly ordered aluminosilicate (AS) networks in zeolites and minerals. The ”B[4] avoidance” analog in BS-based glasses has been questioned,9–11 yet only with ambiguous 3 Environment ACS Paragon Plus


Page 4 of 15

evidence from poorly resolved Raman spectra. In contrast, circumstantial evidence from

11

B/17 O

nuclear magnetic resonance (NMR) experimentation merely suggest that B[4] –O–B[4] bridges are absent in (alumino)borosilicate glasses incorporating Na+ or other network modifiers with low cation field strength (CFS),17–19, 23–26 whereas it has been proposed that B[4] –O–B[4] linkages are stabilized in glasses incorporating cations with high(er) CFS, such as Ca2+ .18, 25–28 Yet, unambiguous proof for B[4] –O–B[4] bonding only exists in NBO-rich Na/Ca-bearing borophosphosilicate glasses.21 By utilizing double-quantum–single-quantum (2Q–1Q)

11

B NMR correlation spectroscopy14, 29–31 supported

by atomistic molecular dynamics (MD) simulations, we resolve these ambiguities/speculations by providing direct and unambiguous experimental evidence that B[4] –O–B[4] contacts are abundant structural motifs in Na2 O–(CaO)–B2 O3 –SiO2 glasses. (insert Table 1 here) Table 1 collects our Na2 O–B2 O3 –SiO2 (NK–R) and Na2 O–CaO–B2 O3 –SiO2 (NCK–R) glass compositions, which employ the {K, R} parameter notation introduced by Bray and co-workers for [3]

RNa2 O–B2 O3 –KSiO2 glasses.8 Table 1 also lists the NMR/MD-derived fractional populations {xB , [4]

[3]

[4]

xB } of {B[3] , B[4] } coordinations (obeying xB +xB = 1) and that of NBO ions (xNBO ) out of the total [4]

O speciation. Section S2 highlights the good agreement between the experimental and modeled {xB , [4]

xNBO } data. In the following, we demonstrate that the xB and xNBO parameters primarily control the B[p] –O–B[q] populations in the glass structure, where the degree of violation of ”B[4] avoidance” is promoted by high NBO contents and the presence of divalent Ca2+ cations. Consequently, the glasses [4]

in Table 1 were selected to arrange wide parameter-spans of 0.27≤ xB ≤0.73 and 0≤ xNBO ≤0.40 with a small set of samples from the two Na2 O–B2 O3 –SiO2 and Na2 O–CaO–B2 O3 –SiO2 systems. Notably, with xE denoting the molar fraction of element E, all glasses herein readily obey


Page 5 of 15 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


1 xSi xO xNBO [4] xB ≤ (3 + 4 − ). 7 xB xB

(1)

For larger B[4] populations, BO4 –BO4 linkages must exist in the structure and ”B[4] avoidance” becomes irrelevant. (Eq. (1) assumes that that NBO only binds to Si/B[3] species). A semi-quantitative mapping of the pair-wise B[p] –O–B[q] bonding preferences is offered by 2Q– 1Q correlation

11

B NMR experimentation14, 29–31 that exploits the through-space-mediated

11

B–11 B

dipolar interaction, which depends on the inverse cube of the 11 B[p] –O–11 B[q] distance. The selection of a short double-quantum coherence (2QC) excitation interval (τexc =167 µs) ensures detection of 2QC signals predominantly from direct

11

B[p] –O–11 B[q] linkages21, 31 (see section S3). The 2Q–1Q

NMR spectra of Fig. 1 evidence the presence of all three B[3] –O–B[3] , B[3] –O–B[4] , and B[4] –O– B[4] bonding constellations throughout all investigated Na2 O–(CaO)–B2 O3 –SiO2 glass networks, as revealed by their respective 2QC signals appearing along the vertical spectral dimension around [3]

33 34 44 δ2Q ≈26 ppm, δ2Q ≈13 ppm, and δ2Q ≈0 ppm. These 2Q shifts are sums of the respective δB [4]

and δB shifts of the 11 B[3] and 11 B[4] resonances that appear along the horizontal “1Q dimension”31 (Fig. 1). The spin-3/2 nature of

11

B produces a narrow NMR peak from the symmetric

environments, whereas much broader signals are observed from the

11

11

BO4

BO3 sites in ”ring” (R) and

”non-ring” (NR) constellations12, 13, 18 (indicated by the shaded areas in Fig. 1a). The fraction of B[p] –O–B[q] linkages out of all B–O–B bridges in the structure is denoted xpq B and was estimated by integrating the associated 2D NMR peak (Fig. 1), followed by correction for the the non-uniform 2QC excitation efficiencies among the three B[p] –B[q] pairs (see the SI). Table 34 44 1 lists the triplets {x33 B , xB , xB } obtained by either NMR experiments or MD simulations, which

reveal a good mutual agreement. Figure 2(a, b) contrasts the experimental/modeled B[4] –O–B[4]



Page 6 of 15

Fig. 1. 2Q–1Q correlation 11 B NMR spectra recorded at 14.1 T and 24.0 kHz MAS from the as-indicated (a–c) Na-based NK–R and (d–f) mixed-cation NCK–R glasses, using one completed [S R212 ] sequence29 (τexc = 167 µs) for generating 2QC signals among direct 11 B[p] –O–11 B[q] linkages, as identified in (c). Projections along the 2Q and 1Q dimensions are displayed to the right and at the top of each 2D NMR spectrum, respectively, together with the MAS NMR spectrum recorded directly by single pulses (red trace). Detailed experimental conditions are given in the SI.

populations with the predictions from a statistical {BO3 , BO4 } intermixing using the underlying [3]

[4]

{xB , xB } data. Notwithstanding that B[4] –O–B[4] bonds are disfavored relative to a statistical borate-group intermixing, they are abundant in all glass structures, regardless of the precise glass compositions and/or nature of the glass-network modifier (Na+ , Ca2+ ). Both experiments and MD simulations reveal a domination of B[3] –O–B[4] linkages (Table 1), which are strongly preferred relative to any other network-former pair in the Na2 O–(CaO)–B2 O3 –SiO2 glasses (Table S4). [p] [q] The abundance xpq linkage in the glass increases concurrently with the B of a given B –O–B [p] [q]

product xB xB (Table 1). Yet, while the borate-group intermixing is strongly disordered, it is not


Page 7 of 15 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


11 Fig. 2. (a, b) Fractional populations of B[4] –O–B[4] linkages (x44 B NMR experB ), as obtained from (a) 2Q–1Q [4] iments and (b) MD simulations, plotted against the experimental/modeled B[4] population (xB ) of the glass. The open symbols represent a statistical intermixing of B[p] –O–B[q] bonds, with the x44 B (stat) data calculated from the [4] respective NMR/MD-derived xB fraction (see Table S3). (c, d) Ratio of the NMR/MD-derived x44 B values and their corresponding statistical predictions. The numbers around the data represent the percentage of NBO out of the total O speciation (100xNBO ). The lines connecting data-points only serve as guide to the eye, while vertical lines mark the two pairs of N(C)2.0–0.5 (cyan) and N(C)4.0–1.2 (red) glasses that only differ in their network modifiers. Error bars are only displayed when outside of the symbols.

statistical, as mirrored in the markedly lower (higher) B[4] –O–B[4] (B[3] –O–B[4] ) contacts than those [3]

[4]

predicted from the {xB , xB } values and a non-preferential BOp –BOq interlinking (Table S3; Fig. 34 44 S1). Besides the BO3 /BO4 populations, two counteracting factors govern the {x33 B , xB , xB } values

and rationalize the observed deviations from a statistical B[p] –O–B[q] bond formation: (i ) The distinct affinities of the various network formers F ={B[3] , B[4] , Si} to coordinate NBO species, which increases in the order B[4]

Direct Experimental Evidence for Abundant BO4âBO4 Motifs in

Recommend Documents

Direct Experimental Evidence for Abundant BO4âBO4 Motifs in