Alkali Activated Organogeopolymers With Volumetri - 2024 - Cement and Concrete C

Cement and Concrete Composites 145 (2024) 105336
Contents lists available at ScienceDirect
Cement and Concrete Composites

journal homepage: www.elsevier.com/locate/cemconcomp
Alkali-activated organogeopolymers with volumetric superhydrophobicity

Dongfang Wang a, c, 1, Li He b, c, 1, **, Yongkang Wu c, d, Yucheng Li c, Weiguo Hu e, Tiantian Ma c, f,
Shengmin Luo g, Jinliang Song c, h, Wei Sun c, Guoping Zhang c, *
a
State Key Laboratory of High Performance Civil Engineering Materials, Jiangsu Sobute New Materials Co., Ltd., Nanjing, 210008, China
b
Institute of Fundamental and Frontier Sciences, University of Electronic Science and Techonology of China, Chengdu 610054, Sichuan, China
c
Department of Civil and Environmental Engineering, University of Massachusetts Amherst, Amherst, MA, 01003, USA
d
School of Civil Engineering, Beijing Jiaotong University, Beijing, 100044, China
e
Department of Polymer Science and Engineering, University of Massachusetts Amherst, Amherst, MA, 01003, USA
f
Wuhan Institute of Rock and Soil Mechanics, Chinese Academy of Sciences, Wuhan, Hubei, 430071, China
g
Deaprtment of Civil and Environmental Engineering, Western New England University, Springfield, MA, 01119, USA
h
College of Transportation Engineering, Dalian Maritime University, Dalian, Liaoning, 116026, China
A R T I C L E I N F O A B S T R A C T
Keywords: A novel class of hybrid organogeopolymers with tunable volumetric superhydrophobicity was developed via a
Alkali activation scalable and catalyst-free route. Its synthesis integrates alkali (NaOH) dissolution of metakaolin, hydrolysis of
Geopolymer methyltrimethoxysilane, and hybrid co-condensation of the resulting silanols (Si(OH)4), sodium-aluminols (NaAl
Hybrid condensation
(OH)4), and methyltrihydroxysilanes (CH3Si(OH)3). Results of molecular structure analyses via X-ray diffraction,
Hydrolysis
Micelle
small-angle X-ray scattering, nuclear magnetic resonance, and Fourier-transform infrared spectroscopy validate
Superhydrophobicity the successful generation of an amorphous inorganic backbone consisting of tetrahedral SiO4/2 and NaAlO4/2
units existing in traditional, purely inorganic geopolymers, but with newly added terminal tetrahedral organic
(–O3/2SiCH3) radicals, which perturb the continuous polycondensation, reduce long-range ordering, and intro
duce micro/nano-pores via hydrophobic aggregation and micelle formation. The unique combination of multi
scale pores and nonpolar methyls lining the pores renders the organogeopolymers three-dimensional
superhydrophobicity that can be tuned by varying the T/Q (where T and Q are tri- and tetra-functional siloxanes
respectively) ratio. For the studied organogeopolymers with variable T/Q molar ratios of 0.02–0.11, but fixed Si/
Al and Na/Al molar ratios of 1.70 and 1.00 respectively, the water contact angle increases from 10 to 169◦ , while
the compressive strength decreases from ~9.8 to 0.5 MPa. The material loses its superhydrophobicity upon
heating to ~370◦ C.
1. Introduction formula –(Si–O)z–Al(Na/K)–O– (where z = Si/Al molar ratio) have also

been developed [4–9]. These two cementitious materials share common
In nature, the Earth’s crust is filled with enormously diverse inorganic chemistry fundamentals: alkaline (e.g., Ca(OH)2, NaOH, KOH) activation
aluminosilicates that make up the superstrong, stable, and durable rocks, of amorphous, reactive silicates and aluminates, which leads to the for
the oldest natural material used in the prehistorical Stone age as well as mation of strong and durable cementitious binders, because the
modern construction. As the world’s most widely used manufactured manufacturing of OPC may be treated as a high-temperature, anhydrous
material, ordinary Portland cement (OPC) discovered nearly two centuries alkaline activation process that suppresses and delays the formation of
ago [1,2] forms upon hydration a solid with superior durability, strength, hydrous cementitious end products (e.g., calcium-silicate-hydrates (CSH),
and other mechanical properties comparable to the counterparts of natural calcium-aluminate-hydrates (CAH)), which can then be generated upon
Portland stones [1–3]. More recently, relatively sustainable geopolymers, addition of water. In fact, alkali dissolution of the reactive aluminosilicates
a type of amorphous inorganic aluminosilicates, with a generic chemical is the foremost geopolymerization reaction, followed by condensation [10,
* Corresponding author.
** Corresponding author. Department of Civil and Environmental Engineering, University of Massachusetts Amherst, Amherst, MA, 01003, USA.
E-mail addresses: [email protected] (L. He), [email protected] (G. Zhang).
1
First authors with equal contributions.
https://doi.org/10.1016/j.cemconcomp.2023.105336
Received 28 June 2023; Received in revised form 24 September 2023; Accepted 13 October 2023
Available online 25 October 2023
0958-9465/© 2023 Elsevier Ltd. All rights reserved.
D. Wang et al. Cement and Concrete Composites 145 (2024) 105336
11] of the tetrahedral silanols (Si(OH)4) and Na-aluminols (NaAl(OH)4) as aluminosilicate minerals, upon amorphization via calcination and/or
monomers to form amorphous geopolymeric tetrahedral backbone. To other treatments (e.g., mechanical milling), have been used as feed
date, geopolymers have spurred academic and industrial interests owing to stocks for the production of cementitious binders, including OPCs and
their environmental benignness, low energy consumption, superior me geopolymers, usually via alkaline activation [36–41]. One key reaction
chanical strength, and excellent chemical and thermal durability [12–17]. of alkaline activation is polycondensation, similar to those encountered
In addition to the manufactured metakaolin, many industrial byproducts in the syntheses of polysiloxanes and polyaluminosiloxanes. In terms of
(e.g., rice husk ash), wastes (fly ash, red mud, furnace slag), and even sustainability and energy efficiency, geopolymer synthesis usually oc
naturally occurring soils and geomaterials such as volcanic ashes and lat curs at ambient temperatures, generates no wastes or CO2 emissions, but
erites [18–20] can be the raw materials used for alkali activation. There consumes industrial residues (e.g., fly ash, furnace slag, mine tailings,
fore, geopolymers can be derived from marginal materials and natural and red mud) and hence geopolymer is regarded as a greener, more
deposits and hence can be an economical and greener alternative to sustainable alternative to the OPC [13,15,16,42–44].
traditional Portland cements. To meet the modern societal and infrastructural developments, more
In the world of manufactured materials, on the other hand, poly diversely functional, high-performance cementitious materials with
organosiloxanes [(R1R2SiO)n, where R1 and R2 are organic radicals] or widespread, functional applications are always desired. Here a question
silicones [21–23] with a wide variety of compositions (e.g., various R1 naturally arises: why cannot the aforementioned two dissimilar Si-
and R2 functional groups), properties (e.g., wettability), and consistency bearing materials, one being purely natural, inorganic aluminosilicates
(e.g., oil, grease, rubber, resin) are extremely versatile, which is ach while the other manufactured and organic siloxanes, be forged into
ieved by varying the siloxane backbone structures (e.g., linear, cyclic, novel hybrid cementitious materials that possess advantageous surface
and cross-linked) and organic side groups with or without crosslinking wettability, chemical and thermal durability, and mechanical strength
(Fig. 1a). Further characterized by inexpensiveness and unique chemical from the silicones and aluminosilicates respectively? The underlying
(e.g., weathering and corrosion resistance), physical (e.g., transparency, supposition is that the tetrahedral siloxanes are contained in OPCs,
thermal stability at both high and low temperatures), and electrical geopolymers, and polysiloxanes. As such, it is logical to find the
properties (e.g., electrical insulation) as well as desired functionalities discrepancy and similarity between the abundant, purely inorganic
(e.g., water repellency, surface energy), organosiloxanes, particularly aluminosilicates that lack organic moieties (e.g., –CH3, –CF3,
the polymeric forms or silicones [24,25], have become increasingly –(CH2)n–CH3) and highly functional, widely used organosiloxanes that
important industrial materials since their invention in the early twen contain no aluminum in their molecular structures. Based on their
tieth century [26–28]. To date, they are among the most extensively structural similarity and compositional contrast of these two abundant
used, silicon-based, manufactured materials spanning numerous in materials, a new class of compounds that possess both aluminum and
dustries, such as construction and building materials, automobiles, organic functional groups may be developed.
medical and personal care products, electronics and electrical devices, Since the basic motifs (C–Si–O) in organosiloxanes or silicones (e.g.,
sensors, and coatings/sealants, among others [29,30]. For example, dimethylsiloxane-based chains) can form an interface between inorganic
augmented by the inclusion of organic groups, an exemplary novel silica and organic compounds [45–47], the challenge is how to create
application of polyorganosiloxanes are preceramic organosilicon poly such interfaces that link purely inorganic aluminosilicates to organic
mers [31] that are used as precursors for the manufacturing of ternary compounds, leading to the formation of organogeopolymers (OG). This
silicon oxycarbide (SiOC) ceramics upon thermal decomposition in inert paper presents a novel class of OG prepared by a new, facile synthesis
environments or pyrolysis [32]. Similarly, highly branched poly route integrating alkali dissolution of amorphous aluminosilicates, hy
silsesquioxanes [(RSiO1.5-)n, Fig. 1b] are improved preceramic polymers drolysis of alkoxysilanes [48,49], and heterofunctional hybrid
for the manufacturing of SiOC ceramics, since their ceramic yield upon co-condensation of inorganic silanols and Na-aluminols as well as
pyrolysis is much higher due to the increased crosslinking, when organosilanols, which all occur in ambient conditions. The hybrid
compared with the linear counterparts, polysiloxanes. In parallel, the co-condensation is proceeded by the generation of reactive species (e.g.,
Al-bearing counterparts, aluminosiloxanes (Fig. 1c and d), have also monomers, dimers, and oligomers) through two different reactions: al
been synthesized via the sol-gel approach as organometallic precursors kali dissolution and hydrolysis. The final products are a new class of
to aluminosilicates that have been suggested as refractory and highly diverse, porous OG with three-dimensional (3D) volumetric
corrosion-resistant coatings for metals and other substrates [33,34]. superhydrophobicity. Such a new synthesis route eliminates catalysts or
Such an approach was also used to fabricate mullite (3Al2O3⋅2SiO2) organic solvents, generates little byproducts except a small amount of
ceramic powders and fibers [34,35]. alcohol (R–OH), and requires little energy input (i.e., no heating, high
As the ubiquitous and most abundant constituents of the Earth’s pressure, or inert atmosphere is required). The major feedstocks are
crust, particularly in rocks and soils readily available for exploitation, usually amorphous aluminosilicates (e.g., metakaolin, fly ash, furnace
Fig. 1. Comparison of the molecular structures of organosiloxanes and organogeopolymers (OG): (a) polysiloxanes; (b) polysilsesquioxanes; (c) aluminosiloxanes; (d)
aluminosilsesquioxanes; (e) three sequential reactions, including alkali dissolution, silane hydrolysis, and hybrid co-condensation, for the synthesis of the new OG.
2
slag, and even mining residues such as red mud), which are widely Table 1
available at large volumes and hence inexpensive. The OG’s molecular Chemical compositions of the metakaolin, sodium silicate solution, and meth
structure bearing aluminum and organic functional groups was char yltrimethoxysilane (MTMS) used for the synthesis of the siloxane (unit: wt.%).
acterized and validated by X-ray powder diffraction (XRD), small angle Composition Metakaolin Na-silicate solution MTMS
X-ray scattering (SAXS), solid state nuclear magnetic resonance Clay-sized (≤2 μm) fraction 38.30
(ssNMR), electron microscopy, and Fourier transform infrared spec Silt-sized (2–75 μm) fraction 58.90
troscopy (FTIR), along with heating treatments. Although the strength of Sand-sized (≥75 μm) fraction 2.80
the new materials is not ideal or practical, such functional cementitious SiO2 56.61 26.5 42.8
Al2O3 39.16
materials, once reinforced for improved strength, can find ample func
Fe2O3 1.87
tional applications in construction. The novelty of this work is man MgO 0.09
ifested by two innovations: (1) the integration of alkali activation, silane CaO 0.05
hydrolysis, and hybrid co-condensation to create a new synthesis route; SO3 0.05
and (2) the grafting of functional organic moieties to the conventional K2O 0.30
Na2O 0.01 10.6
geopolymer backbones, bringing additional functionalities to the
geopolymer-based cementitious materials and composites. H2O 0.13 62.9
LOIa 0.19
2. Materials and methods Total 98.46 100.0 42.8
Data were provided by the manufacturers or suppliers.

2.1. Synthesis of organogeopolymers a
LOI = loss on ignition.
Prior work concluded that the purely inorganic, metakaolin-derived continuous mixing to promote the hydrolysis of MTMS droplets. The
geopolymers synthesized with a nominal Si/Al and Na/Al molar ratio of final homogenized precursor sol was then cast into four customized half-
~2.0 and 1.0 respectively possess relatively higher, optimal unconfined split cylindrical high-density polyethylene (HDPE) molds of 2.54 and
compressive strengths (e.g., ~32–72 MPa) [50,51]. As such, the new 6.35 cm in inner diameter and height respectively, which were pre-
OGs were pre-designed to have a fixed nominal Si/Al, Na/Al, and coated with a very thin layer of silicone-based high vacuum grease on
H2O/Na2O molar ratio of 1.7, 1.0, and 17.5, respectively. The nominal the inner surfaces to facilitate subsequent demolding. The cast precursor
ratios were defined on the basis of the total concentration of each sol samples were initially cured inside the molds at room temperature
respective element, no matter whether it was contained in reactive or for 7 d or until they became stiff enough for handling, followed by
nonreactive minerals. For instance, as a naturally occurring geomaterial, demolding and further curing in air for at least additional 7 d or until
kaolin usually contains some trace impurities such as quartz and they were dried. The setting time was not accurately measured. How
anatase, in addition to the main constituent, kaolinite. Upon calcination ever, observations found that the OG samples with higher T/Q ratios
or dehydroxylation, crystalline kaolinite is converted to amorphous took longer time to harden and dry, and a too large T/Q ratio (e.g.,
reactive metakaolin, while crystalline quartz remains as a nonreactive >0.2–0.3) led to a gel-like material that might take 4–5 weeks to dry or
impurity. Since most chemical analyses such as X-ray fluorescence do might not harden at all. Such a phenomenon warrants further
not differentiate the Si content from metakaolin or quartz, no special investigation.
effort was made to separate the Si contained in nonreactive quartz from For each T/Q ratio, four replicate OG samples were prepared: three
the counterpart in reactive metakaolin. of the final hardened and dried OG samples were used for subsequent
The Si content in the Si/Al ratio included both trifunctional (T = mechanical testing, and the fourth one was cut into smaller sections for
RSiO3/2, where R is an organic hydrocarbon radical) and tetrafunctional molecular and microstructural characterization. Fig. 2a–c show some
(Q = SiO4/2) siloxane units within the OG. Based on preliminary trials, selected pictures of the final hardened OG samples. Also observed was
six different T/Q molar ratios of 0, 0.02, 0.03, 0.04, 0.0625, and 0.11 the self-cracking of some OG samples during the curing and drying
were determined to cover a range of water contact angles. While the T process, indicating that some OG samples have low or little strength,
units were inherited from the methyltrimethoxysilane (MTMS, CH3Si which is inherent to the materials with low surface energy. As discussed
(OCH3)3, AcroSeal™, ACROS Organics™, 97%) via hydrolysis, the Q later, the self-cracking is caused by the drying-induced shrinkage as well
units were directly derived from the inorganic amorphous metakaolin as the materials’ low surface energy. As shown in Fig. 2a–d, no visible
and alkaline activator (i.e., Na-silicate). As discussed later, although the cracks were observed on these samples’ surfaces.
measurements of mechanical properties and water contact angle were
performed on all OG samples, only was the one with a T/Q = 0.0625
analyzed for its molecular structure and other physicochemical prop 2.2. Unconfined compression testing
erties. The reason was that this T/Q ratio was in the middle of the range
of the studied T/Q ratios and yielded the OG with both appropriate For each T/Q ratio, three replicate cylindrical OG samples of 2.54
mechanical integrity and excellent superhydrophobicity. and 6.35 cm in diameter and height respectively were used for uncon
The alkaline activator solution was first prepared by dissolving so fined compression testing in an automated GeoJac loading frame
dium hydroxide (NaOH) pellets (≥97% purity quotient, Fisher Chemi (Trautwein Soil Testing Equipment, Inc., TX, USA) with a maximum load
cal™) in deionized water, followed by adding the sodium-silicate capacity of 8918 N and a load resolution of 0.1 N. However, as discussed
(Na2SiO3) solution (reagent grade, Sigma-Aldrich, Inc., Table 1) to the later, the samples with a very low surface energy and hence a large water
NaOH solution at a pre-designed H2O/Na2O molar ratio of 17.5. The contact angle tended to crack during curing. Therefore, for those OG
mixture solution was stirred continuously and thoroughly for 5 min, samples with higher T/Q ratios, it might be likely that only two samples
followed by cooling to room temperature. The as-received dry powder of were tested. The load cell and the deformation sensor were calibrated
PowerPozz™ metakaolin (MK) (Advanced Cement Technologies, LLC, prior to testing. To minimize the sample-end platen friction and hence
USA, Table 1) was added to the activator solution, followed by stirring the end effect, a thin layer of silicone vacuum grease was applied to the
for at least 15 min to facilitate the alkali-induced dissolution, resulting sample’s top and bottom surfaces that were in contact with the loading
in a homogeneous precursor sol. Then MTMS at the pre-designed T/Q platens. The load was monotonically increased at a constant strain rate
ratios was added to the precursor sol, followed by at least 15 min of 0.5%/h until the specimen failed due to cracking and fracturing.
3
Fig. 2. Pictures showing selected cured OG samples for mechanical testing: (a) to (c) completely cured OG samples with a T/Q ratio of 0.0625, 0.03, and 0.02
respectively. A water droplet on the top surface (in (a) and (b)) does not seep into the sample, indicating its hydrophobicity; (d) a T/Q = 0.02 sample is set up for
unconfined compression testing; (e) a T/Q = 0.04 sample fails under compression in a brittle model; (f) the failure surface of a T/Q = 0.02 sample.
Whenever possible, results averaged from three replicate samples were (WAXS) experiments were performed at the Beamline 9-ID-D of the
reported as the representative compressive strength. Advanced Photon Source, Argonne National Laboratory, using the
grazing incident USAXS/SAXS/WAXS instrument and X-rays with an
2.3. X-ray diffraction and small-angle X-ray scattering (SAXS) energy of 21 keV (or wavelength l = 0.5895 Å) [55,56]. The combined
scattering vector q ranged from 1.0 × 10− 4 to 6 Å− 1 (here q = 4psin(q)/l,
Analyses of the mineralogical compositions of the synthesized OG as and q is half of the scattering angle). The X-ray photon flux was about
well as the source material, metakaolin, were performed in a PAN 5.0 × 1012 mm− 2s− 1. Combined USAXS/SAXS/WAXS measurements
alytical X’Pert PRO X-ray diffractometer (Almelo, Netherlands) using Cu had a time resolution of 5 min. The same dry powder samples prepared
Kα radiation, a scan range of 2◦ –64◦ 2θ (where 2θ is the diffraction above were press-mounted into the Society of Automotive Engineers
angle) and a scan speed of 1◦ 2θ/min. The completely cured and dried (SAE) Type A narrow flat washers made of low carbon steel and with
OG samples were first crushed and pulverized in a percussion mortar dimensions of 6.35 × 15.875 mm (1/4 × 5/8 inch) in thickness and
and pestle (The McCrone Group, Westmont, IL, USA) to pass through a outer diameter respectively. The Scotch magic tape was used as the
0.4 mm opening mesh and then wet ground in a McCrone XRD mill (The window screen to seal the pressed powder. The analyses and modeling of
McCrone Group, Westmont, IL, USA) for 10 min using Propan-2-ol SAXS data were performed by Irena [55–57], a tool suite for analyzing
(C3H7OH) as the grinding liquid to minimize potential crystal struc small-angle neutron scattering and SAXS data within the software Igor
ture damage, resulting in a powder slurry with particle sizes of <4.0 μm Pro (WaveMetrics, Inc., OR, USA).
[52]. The slurry was oven-dried at 105–110 ◦ C for 24 h, and then gently
hand-ground by agate mortar and pestle to pass all materials through a 2.4. Electron microscopy
#625 mesh with an opening size of 20 μm, resulting in a fine powder for
XRD analyses. The razor-tamped surface (RTS) method that can mini The microstructure or micromorphology of the OG and metakaolin
mize preferred orientation of platy particles such as clay minerals was was examined in an FEI Magellan 400 XHR scanning electron micro
adopted to prepare the random powder mounts [53]. For both qualita scope (SEM) (FEI Company, Hillsboro, OR, USA) operated at room
tive and quantitative analyses, a 10 wt% cerianite (CeO2) as an internal temperature and 5 or 10 kV accelerating voltage depending on the
standard was thoroughly mixed to the analyzed sample powder. The selected magnifications to avoid sample damage. All samples were
computer program, Profex, with the Rietveld refinement method [54], examined directly without coating. The nanoscale structure of the OG
was used to analyze all acquired diffraction patterns for phase identifi was also examined and imaged in a JEOL JEM-2200FS energy-filtered
cation and quantification. TEM (TEM) featuring an in-column electron energy loss filter (JEOL
Absolutely calibrated ultrasmall-angle X-ray scattering (USAXS), USA, Inc., Peabody, MA, USA). The prepared powder samples were
small-angle X-ray scattering (SAXS), and wide-angle X-ray scattering dispersed in acetone and a few drops of suspension deposited onto a
4
#300 copper mesh and then air-dried prior to placement into the TEM sandpapers to obtain a relatively flat surface. All measurements were
sample chamber for observation and imaging. performed at room temperature and atmospheric pressure. In addition,
to check the 3D volumetric superhydrophobicity, simple visual checking
2.5. Nuclear magnetic resonance (NMR) spectroscopy was conducted by observing the formation and quick sliding runoff of
nearly spherical water droplets dispensed via a syringe needle on many
All solid-state NMR measurements were conducted on the meta freshly fractured, differently oriented cross-sectional surfaces.
kaolin and OG dry powders at ambient temperature in a Bruker Avance
III HD 600 MHz spectrometer, operating at 1H, 27Al, 13C, and 29Si fre 2.8. Non-reactive molecular dynamics simulation
quencies of 600.13, 156.38, 150.90, and 119.23 MHz, respectively, in a
4.0 mm cross-polarization and magic angle spinning (CP/MAS) broad The 3D geometry and morphology of relatively smaller, individual
band observing probe. The as-received metakaolin powder was used oligomers (i.e., those with 10 and 40 Si atoms) of the OG, as well as the
directly without further processing, while the OG powders were pre aggregation behavior of 36 10-Si-atom oligomers, were investigated by
pared using the same methods described above. To examine the stabil non-reactive molecular dynamics simulations using the molecular me
ity, the OG powder was also heated in a Thermo Scientific Thermolyne chanics calculation engine MM2 embedded within the ChemDraw
benchtop muffle furnace (Barnstead, Inc.) to 400 ◦ C in air for at least 24 platform (PerkinElmer Informatics, Inc., MA, USA). The purpose was to
h, followed by NMR characterization. probe the 3D morphology, formation via hydrophobic attraction of
Specifically, the following experiments were conducted: (1) 1H micelles with hydrophobic groups lining the pores, oligomer folding and
single-pulse experiments with 90◦ excitation, 2 dummy scans, and 8 aggregation, and hence 3D superhydrophobicity. The molecular struc
scans to estimate the 1H relaxation time T1. The 1H T1 was estimated to tures with each atom having a pre-defined number of bonds but bearing
be ~0.5 s for all the samples. The spinning speed was 7–9 kHz; (2) 13C zero charge were first pre-drawn in ChemDraw 2D module and then
and 29Si CP/MAS experiments, where the pulse sequence was a 90◦ pulse directly copied to ChemDraw 3D, followed by energy minimization
followed by a CP period, then 13C detection with high-power 1H using the MM2 calculation engine to optimize the initial geometry of
decoupling. Recycle delay, contact time, decoupling field, and spinning various oligomers. Because the Na+ cations in the oligomers were
speed were 2 s, 2–3 ms, 74 kHz, and 7–8 kHz, respectively; (3) 29Si direct exchangeable and interacted with others via ionic but not covalent
polarization (DP) MAS experiments. Recycle delay, decoupling field, bonds to compensate the deficit charge of tetrahedral AlIV, they were not
and spinning speed were 3–100 s, 74 kHz, and 7–8 kHz, respectively. explicitly considered in these calculations. In this step, the position of
The spectra were found to be completely relaxed at a recycle delay of 3 s. each atom was adjusted so that the cumulative steric energy between
The driving mechanism of the fast 29Si T1 relaxation was likely the unbonded atoms was minimized. However, the interactions between
paramagnetic impurities in the materials; (4) 27Al single-pulse experi bonded atoms, van der Waals attraction, and electrostatic attraction and
ments, with a 10◦ excitation pulse, recycle delay of 2 s, and spinning repulsion between non-bonded atoms were not considered during the
speed of 7–9 kHz. energy minimization. This step basically accelerated the subsequent MD
Deconvolution of the 29Si DP MAS NMR spectrum was conducted simulations so that the simulated systems could reach the equilibrium
using the computer program PeakFit (Systat Software, Inc., USA) to much faster. Then the MD simulation via the MM2 was performed to
differentiate the Q4(cAl) (where c = 0, 1, 2, 3, 4) and T3(dAl) (where d = determine the final configuration, morphology, and aggregation of the
0, 1, 2, 3) molecular clusters in the OG. The fitting process used a oligomers with an equilibrated minimum energy. If the final energy
traditional Gaussian model for each spectroscopic peak and a maximum equilibrium couldn’t be reached, the computation was repeated. The
of 10,000 iterations. The height (i.e., intensity) and linear width (i.e., modified version of Allinger’s MM2 force field embedded with the
standard deviation) of all peaks were allowed to change freely and Chem3D platform was used without modification. Finally, all of the
independently through the fitting and optimization process. Such a above computations used a target temperature of 300 K and a fixed time
fitting process led to the determination of the location, height, and area interval of 2.0 fs, and the typical simulated time duration was 300,000 fs
of all deconvoluted peaks. in total.
2.6. Fourier transform infrared (FTIR) spectroscopy 3. Results and discussion
FTIR reflectance spectroscopy was performed on the same meta 3.1. A new alkali activation process
kaolin and OG powders in a Spectrum 100 FTIR spectrometer (Perkin
Elmer, Inc., USA) at a spatial frequency range of 650–4000 cm− 1. A very The new alkali activation process devised for the OG synthesis con
thin layer of sample powder was loaded into the sample stage, and the sists of three sequential reactions that may not have clearly distin
direct IR beams of the spectrometer were used to radiate the sample to guishable boundaries: alkali dissolution of amorphous reactive
measure the percentage of the beam intensity and at which frequencies aluminosilicates (e.g., metakaolin), hydrolysis of alkoxysilanes (e.g.,
the sample absorbed the infrared light. In addition, to examine the po MTMS), and heterofunctional hybrid co-condensation (Fig. 1e). The first
tential decomposition and oxidation of the organic radicals attached to reaction, validated by the Bayer process for alumina refining [13,44,58]
the OG’s backbone, the powder was also preheated to 200, 300, 350, and recently developed geopolymers [8,13,59,60], occurs under a
370, and 400 ◦ C for at least 24 h in air, followed by FTIR measurements. relatively high alkalinity and results in the formation of a sol consisting
of tetrahedral monomers, including silanol (Si(OH)4) and Na-aluminol
2.7. Water contact angle (WCA) measurements (NaAlIV(OH)4), also termed orthosilicic acid and Na-aluminate respec
tively. A symbolic signature of such a reaction is the transformation of
To examine the OG’s hydrophobicity, the static WCA was measured the pentahedrally coordinated Al (i.e., five-fold AlV) in metakaolin to
in a Dataphysics OCA 15 EC contact angle measurement system (Data tetrahedrally coordinated, four-fold AlIV with an additional Na+ to
Physics Instruments USA Corp., Charlotte, NC, USA). A ~10 μL water balance the charge deficit of an aluminol ([AlIV(OH)4]-), leading to the
drop was dispensed by a syringe needle and static images immediately formation of Na-aluminol, while the silanol is released from the hy
captured and then analyzed. Both finely-ground powder and block droxide ion (OH− )-induced breakdown of the silica tetrahedra (SiO4/2)
samples were examined. The above prepared powder finer than #625 sheets in metakaolin. The second reaction, silane hydrolysis, is a well
mesh (i.e., an opening size of 20 μm) was lightly compacted to form a known one that has been widely exploited by the silicone-based adhe
smooth and flat surface for WCA measurements. Block samples trimmed sives and sealants industry, and the resulting products are methanol
from the original cylindrical samples were manually polished with (CH3OH) and organic methyltrihydroxysilane (CH3-SiT(OH)3, where T
5
denotes a trifunctional siloxane unit) (Fig. 1). Particularly noteworthy is pre-designed Si/Al, Na/Al, and T/Q molar ratios of 1.7, 1.0, and 0.0625
that the high alkalinity required by the first reaction functions as respectively, the corresponding OG is expected to have an ideal formula
catalysis and hence significantly accelerates the hydrolysis reaction [45, [CH3-SiT0.1-O-SiQ
1.6-O-(NaAl)–O-]n (Fig. 1). However, how the tetrahedral
48,49], since the MTMS is initially immiscible in water or the aqueous SiO4/2 and NaAlO4/2 units are linked to form the bulk network is un
hydrophilic sol. In fact, eliminating additional organic solvents and clear. In the following sections, results of analytical characterization of
catalyst is an extra advantage of this new alkali activation process. The the molecular structures of the new OG with a T/Q ratio of 0.0625 as an
third reaction is the hybrid, heterofunctional co-condensation of inor example are used to validate these reactions involved in the new alkali
ganic silanol, inorganic Na-aluminol, and organic methyl-silanol, as well activation process and, more importantly, to uncover the detailed
as the resulting self- and co-condensed dimers, trimers, and oligomers. structure of the bulk OG and its surfaces.
In this reaction, all T units function as terminals hindering further
condensation and hence exit on the surface of the co-condensed tetra 3.2. The amorphous molecular structure
hedral network, while the tetrahedral Q (SiO4/2) and Na-alumina
(NaAlO4/2) units form the internal interconnected backbone as the Occurrence of the above synthesis reactions is first manifested by the
bulk OG’s interior. Particularly, the intermediate oligomers or molecu amorphous molecular backbone of the new OG, which is nearly the same
lar clusters with the nonpolar methyl group as a terminal at one end and as that of the inorganic counterpart, geopolymer. The XRD patterns
polar hydroxyls at the other are amphiphilic and hence tend to form (Fig. 3a) verify and compare the amorphous nature of the source ma
self-assembled micelle-like aggregates with nonpolar methyl groups as terial MK, MK-derived geopolymer, and OG, all with the amorphous
the liners on the internal pore walls. Further noteworthy is that the phase content of ~92.3–94.2 wt% (Table 2). Upon alkali activation, the
hydroxyls on the other end of the molecular clusters can further asymmetrical broad reflection at ~16–40◦ 2θ (2θ = diffraction angle)
condense with others, creating an extended network consisting of from the amorphous MK is converted and shifted to a nearly symmetrical
tetrahedral units bridged by oxygens. one at ~22–36◦ 2θ, which is characteristic of amorphous geopolymers
Ideally, the aforementioned three reactions should occur sequen [7,9,12,60] and also exists with the same shape in OG, validating that
tially to avoid self-condensation of only MTMS, which was achieved by
continuous mixing and by slowly adding the MTMS drop-by-drop. Table 2
However, there is not a clearly definable temporal boundary between Compositions (wt.%) of the MK, geopolymer, and OG (T/Q = 0.0625) analyzed
any two of these reactions, and hence one reaction may occur simulta by semi-quantitative XRD analysis.
neously with the other. For instance, although condensation is ideally Composition Metakaolin Geopolymer OG
desired and actually manipulated to occur after the dissolution of the
Amorphous phase 92.3 94.2 93.9
starting aluminosilicates, partial condensation may have occurred prior
Quartz (SiO2) 6.3 4.4 5.2
to the complete dissolution of all reactive aluminosilicates. Theoreti Anatase (TiO2) 1.4 1.4 0.9
cally, the condensation can deplete all monomers resulting from the Total 98.46 100.0 100.0
alkali activation and silane hydrolysis. For example, according to the
Fig. 3. Amorphous molecular structure of the OG with a T/Q ratio of 0.0625: (a) XRD patterns of the OG, geopolymer, and metakaolin; (b) WAXS spectra of the OG,
geopolymer, and metakaolin; (c) to (e) high resolution TEM micrographs of the OG shoring the amorphous molecular structure.
6
alkali dissolution and condensation, the essential reactions in geo when compared with the much larger aluminosiloxane backbone of the
polymerization, also take place in the OG synthesis. In addition, the geopolymer that shows a continuous, but not aggregated, network
sharp reflections are from two crystalline minerals as impurities, quartz consisting of O-bridged tetrahedra (Fig. 3c and d).
and anatase, in addition to cerianite added as an internal standard for The above amorphous structure is substantially different from the
semi-quantitative XRD analyses. Their presence in all three samples polyorganosiloxanes (POS) or silicones. For instance, the XRD patterns
suggests that these two crystalline phases are not involved in the alkali of polydimethylsiloxane (PDMS) typically exhibit two amorphous peaks
activation reactions, and the geopolymer and OG simply inherit these at ~12.5 and 22.5◦ 2θ, with the former one being relatively sharper, due
nonreactive phases from the source material MK as inert fillers. to the relatively better long-range order of its linear backbone
While little difference can be discerned between the XRD patterns of (–Si–O–Si–) [61,62]. Its other molecular structure features include the
the geopolymer and OG (Fig. 3a), the clear change in the heights of the existence of difunctional D (instead of T in the OG) siloxane units but the
broad peaks at q = ~2.0 and 5.0 Å− 1 in the WAXS spectra (Fig. 3b) does absence of Al and Na. The combination of a high fraction of methyl
reflect different relative crystallinity or long-range disorder of the three groups and a linear siloxane backbone of the PDMS results in a molec
materials. The MK, derived from the calcination of crystalline kaolinite, ular structure and physical and chemical properties that are funda
is expected to preserve to certain degree the repeated ordering of 1:1 mentally different from those of the OG.
layers that consist of one tetrahedral (SiO4)4- and one octahedral ([Al In summary, the new OG and geopolymer have exactly the same
(OH)6]3-) sheets, since dehydroxylation can only occur in the hydroxyl- basic aluminosiloxane network (i.e., –O–Si–O–NaAl–O–) structure that
bearing octahedral sheets. As such, its two broad peaks at q = ~1.5 and is XRD amorphous and is generated via alkali dissolution and conden
4.5 Å− 1 are the tallest and most pronounced when compared with those sation. However, the incorporation of the terminal T units into the OG
of other two materials at respective nearby q. For the OG, the hybrid co- perturbs the continuous growth of the tetrahedra linkage, leading to a
condensation of CH3Si(OH)3 with Si(OH)4 and NaAl(OH)4 introduces an less ordered, smaller amorphous network when compared with
end T unit (CH3–SiTO3/2) and hence terminates the continuous growth of geopolymer.
the aluminosiloxane (–O–Si–O–Al–O) backbone as occurred in the geo
polymer. Therefore, the broad peaks of OG at q = ~2 and 5 Å− 1 are
expected to be smaller or less pronounced than the counterparts of the 3.3. The tetrahedral network
geopolymer (Fig. 3b). Such a small difference cannot be detected by the
conventional XRD but WAXS. High-resolution TEM imaging also vali In general, dehydroxylation of kaolinite (Al2Si2O5(OH)4) leads to the
dates that the OG has an amorphous molecular structure, and its tetra conversion of originally octahedrally-coordinated AlVI to pentahedrally-
hedral SiO4/2 and NaAlO4/2 units tend to form ~10–20 nm aggregates, coordinated AlV in the thermodynamically stable, partially dehydroxy
lated intermediates and fully dehydroxylated metakaolin. Upon alkali
Fig. 4. Magic angle spinning solid state NMR spectra of the OG with a T/Q = 0.0625, 400◦ C-treated OG (OG-400 ◦ C), geopolymer (T/Q = 0.0), and metakaolin: (a)
27
Al direct polarization; (b) 29Si cross polarization by 1H; (c) 1H direct polarization; inset: comparison of the peak positions around 0 ppm; (d) 13C direct polarization.
7
activation, the AlV is further transformed into tetrahedrally-coordinated presence of the organic radicals, -SiCH3. Interestingly, upon heating to
AlIV, one of the two basic tetrahedral units in the backbone of geo 400 ◦ C, the broad peak at 5 ppm of the heating-treated OG becomes
polymer as well as other minerals such as zeolite. Such a conversion can much smaller, and the presence of this residual peak is due to the fact
be readily probed by 27Al MAS ssNMR, which can generally detect all that some surface silanols, (i.e., isolated, geminal, and vicinal silanols)
three types of Al atoms (i.e., AlIV, AlV, and AlVI) found in most materials. can be stable at high temperatures of up to 400–900 ◦ C, or the dehy
As shown in Fig. 4a, the 27Al ssNMR single-pulse spectrum of metakaolin droxylated surface siloxanes can rehydroxylate to form silanols again
differs drastically from those of geopolymer and OG. The former mainly [67]. The other very sharp peak at − 0.15 ppm, but not exactly 0 ppm
consists of the pentahedrally-coordinated AlV, characterized by a broad (Fig. 4c inset), is caused by a similar X–SiCH3 unit, but the X structure in
peak centered at ~30 ppm. This broad peak is superimposed by two very the heating-treated OG is different from the counterpart in the non
small peaks: one at ~0 ppm for trace AlVI, which is a narrow resonance treated one, most likely due to sample contamination. In fact,
signal linked to the crystalline hexa-coordinated AlVI, and the other at silicone-based vacuum grease was used as the lubricant to facilitate
~60 ppm for trace AlIV. The AlVI is usually present in crystalline phases sample demolding, which may be incidentally present in this
such as clay minerals and hence its presence results from the incomplete heating-treated OG sample.
dehydroxylation of kaolinite during calcination. On the other hand, the Similarly, the peak near 0 ppm in the 13C CP NMR spectrum (Fig. 4d)
tetrahedra-coordinated framework aluminum AlIV, as reflected by a is characteristic of -SiCH3 units. Upon heating to 400 ◦ C, this peak di
small, broad, and asymmetric peak, is due to primarily the partially minishes, indicating the thermal decomposition of the methyl group
dehydroxylated intermediates of metakaolin [63,64], and maybe (–CH3) caused by heating. The broad signal between 30 and 90 ppm that
secondarily the minor isomorphous substitution of Si by Al in the are present in both the nontreated and heating-treated OG may be from
tetrahedral sheets of the 1:1 kaolinite layers [65]. For the geopolymer the probe background or some complex carbons single- or double-
and OG, the highest and sharpest peak at ~60 ppm is from the tetra bonded with oxygens. For instance, Na2CO3 or NaHCO3 is widely
hedral AlIV, and the two smaller peaks on both sides are the spinning found as an efflorescing impurity in synthesized geopolymers, as a result
sidebands of this highest peak. Nevertheless, no more pentahedral AlV is of the reaction of atmospheric CO2 with the residual NaOH left from the
discernible, and the crystalline AlVI peak is probably masked by the alkali activation due to incomplete consumption. It is reasonable to as
sideband. In summary, alkali dissolution has completely converted the sume that the OG may contain some residual NaOH and hence Na2CO3
pentahedral AlV to tetrahedral AlIV, and the geopolymer and OG possess or NaHCO3.
nearly the same tetrahedral Al units (i.e., NaAlO4/2) in their molecular
network.
Finally, for the 400 ◦ C-treated OG, no clear distinction is observed in 3.5. Deconvolution for molecular clusters
the spectra of the geopolymer, OG, and the heating-treated OG, sug
gesting that the heating has little influence on the Al-based tetrahedral The above analyses validate that the molecular structure of the OG
units. As discussed later, heating to 400 ◦ C destroys the OG’s super consists of a 3D network of O-bridged tetrahedra, with each consisting of
hydrophobicity, further validating that the T units existing on the sur four O atoms surrounding a silicon or aluminum cation. However, it is
face, but not the Al-tetrahedra units, are one of the origins for the unclear how the SiO4/2 and NaAlO4/2 tetrahedra are quantitatively
superhydrophobicity. Fig. 4b compares the 29Si CP ssNMR spectra of the partitioned in this network, although the maximum Al/Si ratio in such
same four materials. Clearly, the MK has no obvious peaks, indicating tetrahedral networks was established to be 1.0 [68,69]. Fortunately,
that there is no nearby 1H around the tetrahedral Si in the MK, since the deconvolution of the NMR spectra can usually yield quantitative data on
29
Si obtains the CP signal from nearby 1H. This is exactly the case for MK the number and percentage of these relatively longer-range molecular
since dehydroxylation via calcination basically eliminates nearly all clusters consisting of more than one tetrahedral unit. As such, the 29Si
hydroxyls in the original kaolinite, including the octahedral [Al(OH)6]3- DP MAS ssNMR spectrum of the OG was further analyzed by statistical
and broken edge defects such as silanols and aluminols [66]. In contrast, deconvolution (Fig. 5), which is dominated by two relatively broad and
the OG and geopolymer have strong peaks: while the former has two intense peaks at − 128 to − 72 and − 72 to − 44 ppm, corresponding to the
peaks (i.e., at ~− 70 and − 90 ppm for T and Q respectively), only is the Q4 and T3 units respectively.
Q peak found in the latter. Moreover, the 400 ◦ C-treated OG loses the T According to the deconvolution results (Table 3), the signal of the Q4
peak, but keeps the Q peak, suggesting that heating to 400 ◦ C damages units is composed of six deconvoluted subpeaks: 109.83 (Q4(0Al)),
the T siloxane units only. − 101.93 (Q4(1Al)), − 90.89 (Q4(2Al)), − 85.12 (Q4(3Al)), − 77.67
In summary, the synopsis of all 27Al and 29Si NMR spectra of the
three materials clearly validates that OG and geopolymer have the same
tetrahedral NaAlO4/2 and SiO4/2 units, but the former also has extra T
units (i.e., CH3SiTO3/2), which can be destroyed via heating to 400 ◦ C.
Such a comparison of the molecular structures of the three materials
further proves that alkali dissolution, hydrolysis, and co-condensation
reactions do take place in the new alkali activation process, leading to
a complex O-bridged tetrahedra network as the OG’s molecular
backbone.
3.4. Incorporation of organic radicals
The heterofunctional co-condensation of the T units into the inor

ganic network of tetrahedral SiO4/2 and NaAlO4/2 units is also validated
by the 1H and 13C MAS ssNMR spectra (Fig. 4c and d). The lack of hy
droxyls and hence H in the metakaolin whose spectrum only shows the
probe background (Fig. 4c) is due to the high-temperature dehydrox
ylation. Both the geopolymer and OG exhibit a broad peak at 5 ppm,
which is from the silanols (Si–OH), usually as the end groups of the
siloxane backbones or as the defects in the aluminosiloxane network Fig. 5. Deconvolution of the direct polarization 29Si ssNMR spectrum to
[66]. However, the small peak at 0 ppm, unique to the OG, reflects the identify different clusters in the OG with a T/Q = 0.0625.
8
Table 3 cluster violates the rule by having a localized Al/Si ratio of 3/2, which
Quantitative percentages by mole of different Q4(cAl) (c = 0, 1, 2, 3, 4) and exceeds the maximum of 1.0. These nonpolar clusters CH3T3(dAl)
H3CT3(dAl) (d = 0, 1, 2, 3) molecular clusters in the OG (T/Q = 0.0625). (where d is an integer equal to 0 to 2) line the internal pore surfaces,
Q4(cAl) (c = 0, 1, 2, 3, 4) T3(dAl) clusters (d = 0, 1, 2, 3) rendering the porous OG 3D superhydrophobicity, which is to be further
Cluster δ (ppm) Percentage Cluster δ (ppm) Percentage
discussed later.
(%) (%)
Q4(0Al) − 109.83 8.8 T3(0Al) − 63.75 50.9 3.6. FTIR and thermal stability
Q4 in − 107.00 0.7 T3(1Al) − 57.03 11.2
crystals
The nature of chemical bonds in OG and its thermal stability can be
Q4(1Al) − 101.93 32.1 T3(2Al) − 53.14 37.9
4
Q (2Al) − 90.89 32.8 T3(3Al) 0 further analyzed and validated by the FTIR results (Fig. 6). The broad
Q4(3Al) − 85.12 24.5 Total 100.0 peak in the region of 3000–3700 cm− 1 is due to the O–H stretching vi
Q4(4Al) − 77.67 1.1 bration, which is somewhat affected by the acid strength, confinement,
Total 100.0 and location in the framework structure. In general, the characteristic
stretching frequency of an OH group decreases with the number of
(Q4(4Al)) ppm, and a sharp peak for Q4 in crystals at − 107 ppm, of hydrogen bonds. It appears that this broad peak may result mainly from
which the last one is likely to be the signal of crystalline quartz, an the hydrogen-bonded Si–OH (~3500 cm− 1), Al–OH, and hydroxyls (also
impurity directly inherited from the source material, metakaolin. The referred to as bridging hydroxyls) in the Bronsted acid sites (3650-3600
three clusters, Q4(1Al), Q4(2Al), and Q4(3Al), are the majority Q4 cm− 1), the latter being a kind of defects (i.e., Si–O(H)–Al, where H+
groups, accounting for ~90% of all Q4 clusters. Prior XRD analyses also balances the charge deficit of the tetrahedral AlIV) [66]. The broadness
identify and quantify the presence of quartz in the metakaolin and OG of this peak indicates that isolated and germinal (3745 cm− 1), vicinal
(Fig. 3 and Table 2). Similarly, the signal of the T3 units consists of three (3740-3730 cm− 1), and internal (3720-3710 cm− 1) silanols with weak
deconvoluted subpeaks: 63.75 (T3(0Al)), − 57.03 (T3(1Al)), and − 53.14 hydrogen bonds may exist in the OG [71–73]. The occurrence of silanols
(T3(2Al)) ppm, of which the T3(0Al) is the most intense one and accounts and aluminols in the OG is to be discussed later. Moreover, this peak
for 50.9% of all T3 clusters. The two smaller peaks centered around − 25 appears the same for both the geopolymer and OG, despite that the
and − 160 ppm are spinning sidebands of the tallest Q peak at − 90 ppm, former is a hydrophilic material. Another peak that is related to hy
and these sidebands were treated as background during the deconvo droxyls and appears only in the OG and geopolymer is at 1642 cm− 1, due
lution analysis. In any case, these T3 clusters further validate the co- to the O–H bending vibration. The asymmetric stretching of the O–C–O
condensation of the CH3Si(OH)3 with the silanol and Na-aluminol bond is shown at 1420 cm− 1, again due to the presence of Na2CO3 and
generated by alkali dissolution. Finally, the T/Q ratio based on the Na2HCO3 resulting from the reaction of the residual NaOH with atmo
deconvolution result is 0.075, which is slightly greater than the designed spheric CO2 [20].
one, 0.0625. The error may be induced by the presence of non-reactive A sharp and strong peak at 975 cm− 1 is assigned to the M–O–M (M =
quartz, non-calcined residual kaolinite, and other impurities. Al or Si) asymmetric stretching vibration, whereas the sharp band at
The alternating growth mechanisms between tetrahedral SiO4/2 and 1150 cm− 1 can be attributed to the M–O–M in layered metakaolin. A
NaAlO4/2 may basically control the heterofunctional co-condensation, characteristic tiny peak at 1275 cm− 1 is from the Si(C–H3)x stretching
since molecular Al and Si precursors, aluminol and silanol, are prefer vibration. This peak remains in the OG that was heating-treated up to
entially grafted on the silica and alumina surfaces respectively, which 360 ◦ C, but not the 370–400 ◦ C (Fig. 6b), indicating that the methyl
favor the formation of AlIV and SiQ interfacial units [70]. Also, the groups connected to the SiT can remain stable to 360 ◦ C, which is
short-range ordering of framework cations Si and Al in aluminosilicate consistent with the NMR spectra of the 400◦ C-treated OG.
glasses or crystals of tetrahedral aluminates and silicates follows the Al
avoidance or Loewenstein’s rule [68,69], which is expectedly applicable 3.7. Superhydrophobicity and surface morphology
to this OG. As such, the Q4(1Al) and Q4(2Al) accounts for the largest
proportions of all Q4(cAl) (where c is an integer equal to 0 to 4) clusters One of the key functionalities of the newly synthesized OG is their 3D
that form the bulk interior. In contrast, the terminal cluster T3(3Al) is superhydrophobicity (Fig. 7). The pulverized T/Q = 0.0625 OG powder
absent, which may also be due to the Al avoidance rule, since a T3(3Al) of 20–75 μm in particle size has a WCA of ~162.6◦ (Fig. 7a and d), while
Fig. 6. FTIR spectra of: (a) comparison among the OG (T/Q = 0.0625), OG-400 ◦ C, geopolymer (T/Q = 0.0), and metakaolin; (b) comparison among the OG samples
subject to 200–400 ◦ C heating treatments.
9
Fig. 7. Superhydrophobicity of the OG with a T/Q = 0.0625: (a) to (c) water contact angles for the OG powder, fractured chip, and 400◦ C-treated powder,
respectively; (d) and (e) pictures showing the water droplets; (f) schematic illustration of the origin of the OG’s superhydrophobicity.
a WCA of 165.4◦ (Fig. 7b and e) is found on the majority (if not all) of further enhances the hydrophobicity. While above NMR and FTIR re
fractured internal surfaces, suggesting that the OG possess 3D volu sults validate the existence of terminal –SiTCH3 groups lining the pore
metric superhydrophobicity. In principle, a superhydrophobic surface walls, the hierarchical surface roughness can be clearly observed via
defined by a WCA of >150◦ results from two important features: (1) the SEM imaging (Fig. 8). For comparison, the MK and geopolymer lacks
surface must possess a low chemical energy, usually due to the presence such a hierarchical surface roughness.
of nonpolar moieties (e.g., –CH3, –CF3, –SiO3, and their long chains), and Moreover, the 3D superhydrophobicity of the OG must also be
(2) the surface has hierarchical fractal roughness, usually at multiple derived from its micro/nano porous microstructure, which then con
length (e.g., from nano to micro and to millimeter) scales [74–77]. In verts upon fracture or pulverization to the multiscale surface roughness
fact, the achievable maximum WCA reported on ideally smooth surfaces or texture. A similar material reinforced by graphene nanoplatelets
is ~120◦ from a manufactured material with its surface covered by tri possesses a total porosity of 33.3–42.5% [79], which well exceeds those
fluoromethyl groups (i.e., –CF3) [78]. The above WCA values validate of conventional concretes and cements (e.g., 9–12% porosity) [80,81]
that the OG’s interior surface is covered by the –SiTCH3 groups that and is comparable to those of fly ash-based geopolymers (e.g., ~25-25%
lower the chemical energy, and possesses hierarchical roughness that porosity) [82,83]. Moreover, most pore sizes are in the range of
Fig. 8. SEM micrographs comparing the surface micromorphology of different materials: (a) geopolymer (T/Q = 0.0); (b) OG (T/Q = 0.0625); (c) OG-400 ◦ C; (d) to
(f) OG (T/Q = 0.0625)’s hierarchical fractal surface texture.
10
~200–1500 nm. The porous nature of the OG results from the fractured surfaces exhibit superhydrophobicity, the hydrophilic effect of
co-condensation of amphiphiles that form molecular or oligomeric ag Si–OH and Al–OH induced by broken bonds on the overall hydropho
gregates such as micelles, bilayers, and liposomes. The condensation of bicity of a fractured surface is not significant, most likely due to the
silanols and Na-aluminols leads to hydrophilic oligomers and polymers, smaller fraction of the Si–OH and Al–OH groups when compared with
while co-condensation of methylsilanols adds a hydrophobic end or the –SiTCH3 units. In fact, the pulverized powder particles of OG may
terminal to the hydrophilic oligomers and polymers, which hence have both hydrophilic and hydrophobic zones on their surface.
become amphiphilic and can then form micelles and liposomes via hy These experimentally observed phenomena can also be validated, at
drophobic aggregation and attraction. Furthermore, since the T units least partially, by the nonreactive MD simulations. First, individual
always act as terminating groups forming a surface with a relatively oligomers formed via the process of co-condensation exhibit an amor
lower chemical surface energy and higher water repellency, such an phous structure with a folded and curved –O–Si–O–Al–O– backbone and
enclosed surface then forms as a pore within the bulk OG. curved rough surface (Fig. 9). Second, when multiple oligomers are
Two additional aspects are further noteworthy. The first one is the interacting with each other via surface and interfacial forces, but not
distinctive low surface energy of the OG, which stems from (1) the condensation, highly complex and porous oligomer aggregates with
organic functional groups (e.g., mostly methyl and aliphatic fluoro highly irregular rough surfaces are formed, as evidenced by the ultra
carbon groups) linked to the silicon atoms and (2) the symmetrical high resolution TEM images (Fig. 10a–c). The aggregates’ external
nonpolar tetrahedral siloxanes without broken Si–O–Si bonds. Although surface and internal pore wall are terminated with nonpolar –SiTCH3
the difference between the electronegativity of the two atoms of the groups that cannot be condensed, and hence are hydrophobic. More
siloxane bond (i.e., 1.8 for Si and 3.5 for O) can results in a stronger bond over, this hydrophobicity is further enhanced by the irregularly curved
having a polar, partial ionic (~40%) character, the tetrahedral SiO4/2 on surface at different scales: (1) the smallest scale is the size of a –CH3
phyllosilicates, such as those on the basal surfaces of natural talc and group (Fig. 10c–e); (2) the intermediate scales result from the backbone
pyrophyllite minerals, usually exhibit hydrophobicity [84–87]. The folding and entangling as well as the micelle-like structure (Fig. 10e); (3)
second one is the effect of polar, hydrophilic Si–OH and Al–OH groups the largest scale is the size of aggregates (Fig. 8). Again, similar to
on fractured surfaces caused by breaking the covalent bonds in the amphiphilic surfactants, the oligomers may form micelles, liposomes, or
backbone structure –O–Si–O–NaAl–O–, as evidenced by the FTIR and other similar porous structures with the pore wall lined by the termi
NMR results (Figs. 4–6). Since both the pulverized OG powder and nating -SiTCH3 groups, which lead to the formation of multiscale,
Fig. 9. OG (T/Q = 0.0625) oligomer’s morphology generated by non-reactive MD simulations: (a) initial 2D structure with 2 SiT, 8 SiQ, and 6 Al; (b) and (c) the
morphology of a 10-Si oligomer viewed at different directions; (d) and (e) the morphology of a 40-Si oligomer viewed at different directions.
11
Fig. 10. Formation of surface roughness and molecular aggregates with nanopores in the OG (T/Q = 0.0625): (a) and (b) TEM micrographs showing the rough
surface and molecular aggregates of ~10–50 nm in size; (c) formation of nanopores and nonpolar rough surface observed via nonreactive MD simulations of the
aggregation of 36 10-Si OG oligomers; (d) aggregation and micelle formation via hydrophobic attraction; (e) continuous condensation of hydrophilic end groups -Si-
OH and -Al-OH to grow aggregates until reaching a hydrophobic counterpart –SiT–CH3 on the surface; (f) resemblance of the OG’s hierarchical surface texture
to cauliflower.
hierarchically-structured surface roughness (Fig. 10c–e). Continuous a micro/nano porous structure that, upon fracture, makes up the rough
condensation and growth of these oligomers eventually lead to the texture. On the other hand, a fractured surface is usually occupied by
formation of a cauliflower-like surface (Fig. 10f), which possesses a hydrophilic moieties (e.g., –OH for most oxides such as hematite and
multiscale, hierarchical surface texture and nonpolar functional groups, silica, and even clay mineral edge surfaces) due to broken bonds
two essential features for superhydrophobicity. [91–93]. According to the Wenzel model [94], surface roughness can
decrease the WCA of a hydrophilic surface but increase it for a hydro
phobic one. As such, the low chemical energy of a fractured surface must
3.8. Tunable superhydrophobicity and strength
be derived from the nonpolar functional groups lining the internal walls
of the micro/nano pores, which can also partially compensate the hy
For the studied range of T/Q ratios of 0.0–0.11, the OG powder has a
drophilicity of the broken bonds on the fractured surface. Accordingly, a
WCA of ~1 (to 169◦ , i.e., from a hydrophilic porous material to a
3D superhydrophobic material usually exhibits very low strength and
superhydrophobic one (Table 4). Also, dramatic change in the WCA
fracture roughness because an internal cross-sectional fracture surface
occurs at T/Q = 0.02 to 0.0625. However, the gain in super
always has the lowest surface energy, or fracture always occur on an
hydrophobicity is sacrificed by the loss in compressive strength: the
internal surface with the lowest surface energy. Such a basic, inherent
geopolymer has a compressive strength of 32 MPa, while the OG with a
physical law explains why the superhydrophobic OG has a low
WCA of ~162–169◦ has a strength of ~0.50 MPa, i.e., a 98% reduction
compressive strength.
of the geopolymer’s strength. In particular, most OG samples still exhibit
a mode of brittle failure (Fig. 2e and f). The unconfined compression
stress-strain curves also manifest the brittle failure of these OG samples 4. Discussion
(Fig. 11). For the purpose of comparison, the effect of the OG’s T/Q
ratios on the compressive strength and WCA is summarized in Table 4. 4.1. Synthesis reactions and versatility
As expected, the WCA or hydrophobicity of the OG varies with the T/
Q ratio (Fig. 12). In general, the most striking feature of 3D volumetric The new synthesis method for the OG integrates for the first time
superhydrophobic materials is their WCA of >150◦ on the majority, if three widely proven reactions: alkali dissolution of calcined alumino
not virtually any, of internal surfaces [88–90]. According to the two silicates, hydrolysis of alkoxysilanes, and heterofunctional co-
requirements for superhydrophobicity, the fractured internal surface condensation, and hence creates a new sol-gel-based pathway for
must possess multiscale surface roughness, which is usually achieved by manufacturing highly diversified functional materials. First, since the
Si/Al ratio of geopolymers can be varied from a minimum of 1 to >15 or
even ~300 [7,60], the hybrid counterparts—OG can expectedly be
Table 4
Effects of the T/Q ratio on the compressive strength and WCA of the OG samples.
synthesized with a similar range of Si/Al ratios. Second, since the
Purely inorganic geopolymer has a T/Q = 0.0 tetrahedral (AlO4/2)- requires an additional monovalent alkali cation (e.
g., Na+, K+, Li+) for charge neutralization, these options for different
T/Q ratio Strength (MPa) WCA (◦ )
alkali cations can further enhance the OG’s diversity. Third, many other
0.11 0.49 ± 0.03 168.97 ± 0.86 hydrolytic alkoxysilanes (e.g., mono-, di-, and tetra-functional silanes,
0.0625 0.5a 166.56 ± 0.10
M, D, and Q) with even more diverse organic moieties (–CF3, phenols,
0.04 0.54 ± 0.02 162.05 ± 0.04
0.03 2.16 ± 0.03 158.60 ± 0.05 and their chains) can also be co-condensed into the geopolymeric
0.02 9.76 ± 0.13 10.0 ± 0.24 backbones at different fractions (e.g., the T/Q and D/Q molar ratios),
0.0 (geopolymer) 32.75 ± 0.93 1.0b further enhancing the versatility of the OG with potentially varied
a
Only one sample was tested for its strength measurement. functionalities and structures. Finally, a wide array of aluminosilicates,
b
For geopolymer, water droplet seeped into its bulk interior and hence its the most abundant minerals on the earth’s crust, can be used after
WCA was assumed to be 1◦ . No standard deviation could be obtained since there calcination, mechanical grinding, and other amorphization techniques
was no droplet formed on the porous surface. as the source materials for the synthesis of OG.
12
Fig. 11. Selected unconfined compression stress-strain curves: (a) purely inorganic geopolymer (T/Q = 0.0); (b) OG with Tvarying T/Q ratios.
of the sacrificed lower strength that may be improved by other tech

niques, as described in the next section, the hydrophobic and super
hydrophobic OG may be used as independent load-bearing structural
members or a protective layer over conventional cementititous struc
tural materials, rendering the structures other attractive functionalities,
such as self-cleaning, anti-icing, anti-corrosion, and anti-biofouling,
among others, as demonstrated by other superhydrophobic surfaces,
coatings, and bulk materials. For example, most reinforced concretes
based on ordinary Portland cement have shrinkage cracks that allow for
water percolation and hence steel rebar corrosion. However, a hydro
phobic surficial layer made of the new OG can eliminate such potential
damages by preventing water seepage even the structural concretes have
cracks.
As pointed out earlier, 3D volumetric superhydrophobic materials
with a low surface energy along many cross-sectional surfaces are
inherently porous and prone to fracture failure. Such an inherent
drawback limits the practical, widespread applications of 3D super
Fig. 12. Dependence of the OG’s hydrophobicity and compressive strength hydrophobic materials. To overcome this deficiency, internal re
upon the T/Q molar ratio. Purely inorganic geopolymer is assumed to have a inforcements that can be mixed or added to the bulk matrix are desired
WCA of zero since its porous structure absorbs water droplet into its interior and relevant technologies should be developed. For the studied OG, a
bulk volume. The two vertical broken lines indicate a narrow range of T/Q WCA of 90–150◦ (i.e., hydrophobic) measured on the ground powder
molar ratios for the OG to be hydrophobic and have certain compres corresponds to an unconfined compressive strength of ~6.0 to 2.0 MPa
sive strength. (Fig. 12). Such low strengths certainly disqualify the materials’ appli
cations as structural cementitious materials. However, a parallel study
When compared with the synthesis of purely inorganic geopolymers, [79] demonstrated that adding 0.9 wt% graphene nanoplatelets
the condensation reaction of the OGs is perturbed by the introduced T improved the compressive strength to ~34.0 MPa, well exceeding the
groups bearing the organic moieties (i.e., –CH3), which act as terminals 28-day strength (i.e., 21 MPa) of Type V Portland cement, yet the
stopping the continuous condensation. As a result, the length or size of composites are still superhydrophobic with a WCA of 156.6◦ . The
the OG chains or networks is reduced, resulting in reduced long-range structural applications of superhydrophobic cementitious materials such
ordering of the hybrid polymers. The degree of such reduction is man as OG discussed in this paper warrants further research.
ifested by at least the two observed properties: (1) the shorter broad
reflection of the OG in the WAXS spectrum (Fig. 2b), and (2) the reduced 5. Conclusions
unconfined compressive strength of the OG, although the latter is more
complicated and affected by many other factors such as porosity, den This paper introduces a novel class of hybrid organogeopolymers
sity, surface energy, and presence of microcracks, among others. Also, as (OG) with tunable volumetric superhydrophobicity. Extensive charac
shown in Fig. 11, the more the T groups in the OG, the lower the terization was conducted to validate and understand the OG’s molecular
compressive strength. One of the underlying mechanisms for such and micromorphological structures, functionalities such as tunable hy
strength reduction may also stem from the reduced polymeric chain drophobicity and strength, and the synthesis principles. The following
length or long-range ordering. Future effort should be made to quanti conclusions can be drawn based on above analyses and discussion.
tatively measure the polymeric dimensions (e.g., chain length) as well as
the molecular weights of the OGs. • A new synthesis route was created to manufacture OG, and it in
tegrates for the first time alkali dissolution of amorphous alumi
4.2. Practical concerns and implications nisilicates, hydrolysis of alkoxysilanes, and hybrid co-condensation
of the reactive monomers, dimers, and oligomers.
The above compositional and structural versatility can be expected • The new synthesis route was validated by the spectroscopic and
to lead to various functionalities and hence diverse applications. In spite molecular structural data.
13
• The new OG are hybrid, functionalized cementitious materials with [10] A. Buchwald, H.D. Zellmann, C. Kaps, Condensation of aluminosilicate gels-model
system for geopolymer binders, J. Non-Cryst. Solids 357 (2011) 1376–1382,
nearly the same continuous network of O-bridged tetrahedra of SiO4/
https://doi.org/10.1016/j.jnoncrysol.2010.12.036.
2 and NaAlO4/2 or aluminosiloxane backbone as the purely inorganic [11] G. Geng, R.J. Myers, J. Li, R. Maboudian, C. Carraro, D.A. Shapiro, P.J.
geopolymers. However, the OG contain organic methyl groups acting M. Monteiro, Aluminum-induced dreierketten chain cross-links increase the
as terminals on the backbone and lining the walls of internal pores mechanical properties of nanocrystalline calcium aluminosilicate hydrate, Sci. Rep.
7 (2017), https://doi.org/10.1038/srep44032.
formed via micelle-like hydrophobic aggregations; [12] P. Duxson, G.C. Lukey, J.S.J. van Deventer, Thermal evolution of metakaolin
• The bulk OG possesses 3D volumetric superhydrophobicity and geopolymers: Part 1 - physical evolution, J. Non-Cryst. Solids 352 (2006)
multiscale pores. While the hydrophobicity increases with the T/Q 5541–5555, https://doi.org/10.1016/j.jnoncrysol.2006.09.019.
[13] A. Van Riessen, E. Jamieson, C.S. Kealley, R.D. Hart, R.P. Williams, Bayer-
ratio, their compressive strength generally shows the opposite trend; geopolymers: an exploration of synergy between the alumina and geopolymer
• The new OG are an addition to the realm of existing cementitious industries, Cem. Concr. Compos. 41 (2013) 29–33, https://doi.org/10.1016/j.
materials and enhance their diversity, versatility, and functionality. cemconcomp.2013.04.010.
[14] X. Ge, L. Duran, M. Tao, D.J. DeGroot, E. Li, G. Zhang, Characteristics of
underwater cast and cured geopolymers, Cem. Concr. Compos. 114 (2020),
Declaration of competing interest https://doi.org/10.1016/j.cemconcomp.2020.103783.
[15] G. Habert, C. Ouellet-plamondon, Recent Update on the Environmental Impact of
Geopolymers, 2016, pp. 17–23.
The authors declare the following financial interests/personal re [16] A.L. Freire, H.J. José, R. de F.P.M. Moreira, Potential applications for geopolymers
lationships which may be considered as potential competing interests: in carbon capture and storage, Int. J. Greenh. Gas Control 118 (2022), 103687,
https://doi.org/10.1016/j.ijggc.2022.103687.
Dongfang Wang reports financial support was provided by Geo [17] N.B. Singh, B. Middendorf, Geopolymers as an alternative to Portland cement: an
synthetics Institute. Li He reports was provided by Geosynthetics overview, Construct. Build. Mater. 237 (2020), https://doi.org/10.1016/j.
Institute. conbuildmat.2019.117455.
[18] R.C. Kaze, A. Naghizadeh, L. Tchadjie, A. Adesina, J. Noel Yankwa Djobo, J.
G. Deutou Nemaleu, E. Kamseu, U. Chinje Melo, B.A. Tayeh, Lateritic soils based
Data availability geopolymer materials: a review, Construct. Build. Mater. 344 (2022), https://doi.
org/10.1016/j.conbuildmat.2022.128157.
[19] R.C. Kaze, G.L. Lecomte-Nana, E. Kamseu, P.S. Camacho, A.S. Yorkshire, J.
Data will be made available on request.
L. Provis, M. Duttine, A. Wattiaux, U.C. Melo, Mechanical and physical properties
of inorganic polymer cement made of iron-rich laterite and lateritic clay: a
Acknowledgements comparative study, Cement Concr. Res. 140 (2021), https://doi.org/10.1016/j.
cemconres.2020.106320.
[20] R.C. Kaze, J.N.Y. Djobo, A. Nana, H.K. Tchakoute, E. Kamseu, U.C. Melo,
Partial supports from the Geosynthetic Institute (GSI) Fellowships C. Leonelli, H. Rahier, Effect of silicate modulus on the setting, mechanical strength
and the Perrell Scholarships of the University of Massachusetts Amherst, and microstructure of iron-rich aluminosilicate (laterite) based-geopolymer cured
both awarded to Drs. D. Wang and L. He, are gratefully acknowledged. at room temperature, Ceram. Int. 44 (2018) 21442–21450, https://doi.org/
10.1016/j.ceramint.2018.08.205.
[21] F. Schneider, T. Fellner, J. Wilde, U. Wallrabe, Mechanical properties of silicones
Abbreviations for MEMS, J. Micromech. Microeng. 18 (2008), https://doi.org/10.1088/0960-
1317/18/6/065008.
[22] M.J. Owen, The surface activity of silicones: a short review, Ind. Eng. Chem. Prod.
FTIR Fourier transform infrared spectroscopy Res. Dev. 19 (1980) 97–103, https://doi.org/10.1021/i360073a023.
HDPE High-density polyethylene [23] W. Noll, Chemistry and Technology of Silicones, Elsevier Science, Oxford, 1968.
MK metakaolin [24] W. Noll, Chemistry and Technology of Silicones, Academic Press Inc, 1968, https://
doi.org/10.1016/B978-0-125-20750-8.X5001-5.
MTMS methyltrimethoxysilane [25] C. Rücker, K. Kümmerer, Environmental chemistry of organosiloxanes, Chem. Rev.
NMR nuclear magnetic resonance 115 (2015) 466–524, https://doi.org/10.1021/cr500319v.
OAS organoaluminosiloxane [26] R.R. LeVier, M.C. Harrison, R.R. Cook, T.H. Lane, What is silicone? J. Clin.
Epidemiol. 48 (1995) 513–517.
OG organogeopolymer [27] H.W. Fox, P.W. Taylor, W.A. Zisman, Polyorganosiloxanes....Surface active
POAS polyorganoaluminosiloxane properties, Ind. Eng. Chem. 39 (1947) 1401–1409.
SEM scanning electron microscopy [28] S.C. Shit, P. Shah, A review on silicone rubber, Natl. Acad. Sci. Lett. 36 (2013)
355–365, https://doi.org/10.1007/s40009-013-0150-2.
TEM transmission electron microscopy [29] P.A. Storozhenko, G.I. Shcherbakova, Synthesis of high-purity aluminosilicates
D, T, Q difunctional, trifunctional, tetrafunctional siloxane units using aluminosiloxane precursors, Inorg. Mater. 47 (2011) 167–171, https://doi.
XRD X-ray powder diffraction org/10.1134/S002016851102018X.
[30] T. Ogawa, J. Watanabe, Y. Oshima, Catalyst-free synthesis of polyorganosiloxanes
WAXS wide-angle X-ray scattering
by high temperature & pressure water, J. Supercrit. Fluids 45 (2008) 80–87,
https://doi.org/10.1016/j.supflu.2007.11.010.
References [31] P. Colombo, G. Mera, R. Riedel, G.D. Sorarù, Polymer-derived ceramics: 40 Years
of research and innovation in advanced ceramics, J. Am. Ceram. Soc. 93 (2010)
1805–1837, https://doi.org/10.1111/j.1551-2916.2010.03876.x.
[1] P.E. Halstead, The early history of portland cement, Trans. N.com. Soc. 34 (1961)
[32] S. Somiya, Handbook of Advanced Ceramics, second ed., Academic Press Inc, 2013
37–54, https://doi.org/10.1179/tns.1961.003.
https://doi.org/10.1016/C2010-0-66261-4.
[2] G.C. Bye, Portland Cement: Composition, Production and Properties, Thomas
[33] A.G. Williams, L. v Interrante, Design and synthesis of metal-organic precursors to
Telford, 1999.
aluminosilicates, in: MRS Online Proceedings Library, 1984, https://doi.org/
[3] A.J. Allen, J.J. Thomas, H.M. Jennings, Composition and density of nanoscale
10.1557/PROC-32-151. (Accessed 27 May 2022).
calcium-silicate-hydrate in cement, Nat. Mater. 6 (2007) 311–316, https://doi.org/
[34] T. Yogo, I.A. Aksaf, Synthesis of mullite fibre from an aluminosiloxane precursor,
10.1038/nmat1871.
J. Mater. Chem. 4 (1994) 353–359.
[4] J. Davidovits, Geopolymers: inorganic polymeric new materials, J. Therm. Anal.
[35] D.R. Treadwell, D.M. Dabbs, I.A. Aksay, Mullite (3Al2O3-2SiO2) synthesis with
Calorim. 37 (1991) 1633–1656, https://doi.org/10.1007/BF01912193.
aluminosiloxanes, Chem. Mater. 8 (1996) 2056–2060.
[5] R.E. Lyon, P.N. Balaguru, A. Foden, U. Sorathia, J. Davidovits, M. Davidovics, Fire-
[36] H. Xu, J.S.J. Van Deventer, The geopolymerisation of alumino-silicate minerals,
resistant aluminosilicate composites, Fire Mater. 21 (1997) 67–73, https://doi.org/
Int. J. Miner. Process. 59 (2000) 247–266. www.elsevier.nlrlocaterijminpro.
10.1002/(SICI)1099-1018(199703)21:2<67::AID-FAM596>3.0.CO;2-N.
[37] P. Duxson, G.C. Lukey, J.S.J. van Deventer, Thermal evolution of metakaolin
[6] P. Duxson, A. Fern, J.L. Provis, G.C. Lukey, A. Palomo, J.S.J. Van Deventer,
geopolymers: Part 1 - physical evolution, J. Non-Cryst. Solids 352 (2006)
Geopolymer technology: the current state of the art, J. Mater. Sci. 42 (2007)
5541–5555, https://doi.org/10.1016/j.jnoncrysol.2006.09.019.
2917–2933, https://doi.org/10.1007/s10853-006-0637-z.
[38] D. Kim, H.T. Lai, G.V. Chilingar, T.F. Yen, Geopolymer formation and its unique
[7] R.A. Fletcher, K.J.D. MacKenzie, C.L. Nicholson, S. Shimada, The composition
properties, Environ. Geol. 51 (2006) 103–111, https://doi.org/10.1007/s00254-
range of aluminosilicate geopolymers, J. Eur. Ceram. Soc. 25 (2005) 1471–1477,
006-0308-z.
https://doi.org/10.1016/j.jeurceramsoc.2004.06.001.
[39] P. Duxson, A. Fern, J.L. Provis, G.C. Lukey, A. Palomo, J.S.J. Van Deventer,
[8] H. Xu, J.S.J. Van Deventer, The geopolymerisation of alumino-silicate minerals,
Geopolymer technology: the current state of the art, J. Mater. Sci. 42 (2007)
Int. J. Miner. Process. 59 (2000) 247–266. www.elsevier.nlrlocaterijminpro.
2917–2933, https://doi.org/10.1007/s10853-006-0637-z.
[9] X. Ge, L. Duran, M. Tao, D.J. DeGroot, E. Li, G. Zhang, Characteristics of
[40] F. Pacheco-Torgal, J. Castro-Gomes, S. Jalali, Alkali-activated binders: a review.
underwater cast and cured geopolymers, Cem. Concr. Compos. 114 (2020),
Part 1. Historical background, terminology, reaction mechanisms and hydration
https://doi.org/10.1016/j.cemconcomp.2020.103783.
14
products, Construct. Build. Mater. 22 (2008) 1305–1314, https://doi.org/10.1016/ [67] L.T. Zhuravlev, The Surface Chemistry of Amorphous Silica, Zhuravlev model,
j.conbuildmat.2007.10.015. 2000. www.elsevier.nl/locate/colsurfa.
[41] J.L. Provis, Alkali-activated materials, Cement Concr. Res. 114 (2018) 40–48, [68] W. Loewenstein, The distribution of aluminum in the tetrahedra of silicates and
https://doi.org/10.1016/j.cemconres.2017.02.009. aluminates, Am. Mineral. 39 (1954) 92–96.
[42] P. Renforth, The negative emission potential of alkaline materials, Nat. Commun. [69] S. Keun, J.F. Stebbins, The degree of aluminum avoidance in aluminosilicate
10 (2019), https://doi.org/10.1038/s41467-019-09475-5. glasses, Am. Mineral. 84 (1999) 937–945.
[43] B.C. McLellan, R.P. Williams, J. Lay, A. van Riessen, G.D. Corder, Costs and carbon [70] M. Valla, A.J. Rossini, M. Caillot, C. Chizallet, P. Raybaud, M. Digne,
emissions for geopolymer pastes in comparison to ordinary portland cement, A. Chaumonnot, A. Lesage, L. Emsley, J.A. van Bokhoven, C. Copéret, Atomic
J. Clean. Prod. 19 (2011) 1080–1090, https://doi.org/10.1016/j. description of the interface between silica and alumina in aluminosilicates through
jclepro.2011.02.010. dynamic nuclear polarization surface-enhanced NMR spectroscopy and first-
[44] E. Jamieson, B. McLellan, A. van Riessen, H. Nikraz, Comparison of embodied principles calculations, J. Am. Chem. Soc. 137 (2015) 10710–10719, https://doi.
energies of ordinary portland cement with bayer-derived geopolymer products, org/10.1021/jacs.5b06134.
J. Clean. Prod. 99 (2015) 112–118, https://doi.org/10.1016/j. [71] X.S. Zhao, G.Q. Lu, A.K. Whittaker, G.J. Millar, H.Y. Zhu, Comprehensive study of
jclepro.2015.03.008. surface chemistry of MCM-41 using 29 Si CP/MAS NMR, FTIR, pyridine-TPD, and
[45] H. Muramatsu, R. Corriu, B. Boury, Solid state hydrolysis/polycondensation of TGA, J. Phys. Chem. B 101 (1997) 6525–6531. https://pubs.acs.org/sharingg
alkoxysilane: access to crystal-like silicon-based hybrid materials, J. Am. Chem. uidelines.
Soc. 125 (2003) 854–855, https://doi.org/10.1021/ja020975e. [72] Z.W. Peng, P.T. Hsieh, Y.J. Lin, C.J. Huang, C.C. Li, Investigation on blistering
[46] R.J.P. Corriu, J.J.E. Moreau, P. Thepot, M. Wong, C. Man, New Mixed Organic- behavior for n-type silicon solar cells, in: Energy Procedia, Elsevier Ltd, 2015,
Inorganic Polymers: Hydrolysis and Polycondensation of Bis(trimethoxysilyl) pp. 827–831, https://doi.org/10.1016/j.egypro.2015.07.117.
Organometallic, 1992. https://pubs.acs.org/sharingguidelines. [73] T.W. Dijkstra, R. Duchateau, R.A. van Santen, A. Meetsma, G.P.A. Yap,
[47] M. Frangoise Surivet, T. My Lam, J.-P. Pascault, Q. Tho Pham, Organic-Inorganic Silsesquioxane models for geminal silica surface silanol sites. A spectroscopic
Hybrid Materials. 1. Hydrolysis and Condensation Mechanisms Involved in investigation of different types of silanols, J. Am. Chem. Soc. 124 (2002)
Alkoxysilane-Terminated, 1992. https://pubs.acs.org/sharingguidelines. 9856–9864, https://doi.org/10.1021/ja0122243.
[48] A.A. Issa, M. El-Azazy, A.S. Luyt, Kinetics of alkoxysilanes hydrolysis: an empirical [74] S. Singh, J. Houston, F. van Swol, C.J. Brinker, Superhydrophobicity: drying
approach, Sci. Rep. 9 (2019), https://doi.org/10.1038/s41598-019-54095-0. transition of confined water, Nature 442 (2006) 526, https://doi.org/10.1038/
[49] S.A. Torry, A. Campbell, A.v. Cunliffe, D.A. Tod, Kinetic analysis of organosilane 442526a.
hydrolysis and condensation, Int. J. Adhesion Adhes. 26 (2006) 40–49, https://doi. [75] X. Yu, Z. Wang, Y. Jiang, X. Zhang, Surface gradient material: from
org/10.1016/j.ijadhadh.2005.03.008. superhydrophobicity to superhydrophilicity, Langmuir 22 (2006) 4483–4486,
[50] D.S. Perera, J.V. Hanna, J. Davis, M.G. Blackford, B.A. Latella, Y. Sasaki, E. https://doi.org/10.1021/la053133c.
R. Vance, Relative strengths of phosphoric acid-reacted and alkali-reacted [76] D. Wang, Q. Sun, M.J. Hokkanen, C. Zhang, F.Y. Lin, Q. Liu, S.P. Zhu, T. Zhou,
metakaolin materials, J. Mater. Sci. 43 (2008) 6562–6566, https://doi.org/ Q. Chang, B. He, Q. Zhou, L. Chen, Z. Wang, R.H.A. Ras, X. Deng, Design of robust
10.1007/s10853-008-2913-6. superhydrophobic surfaces, Nature 582 (2020) 55–59, https://doi.org/10.1038/
[51] J. He, J. Zhang, Y. Yu, G. Zhang, The strength and microstructure of two s41586-020-2331-8.
geopolymers derived from metakaolin and red mud-fly ash admixture: a [77] J.T. Simpson, S.R. Hunter, T. Aytug, Superhydrophobic materials and coatings: a
comparative study, Construct. Build. Mater. 30 (2012) 80–91, https://doi.org/ review, Rep. Prog. Phys. 78 (2015), https://doi.org/10.1088/0034-4885/78/8/
10.1016/j.conbuildmat.2011.12.011. 086501.
[52] A.J. Locock, D. Chesterman, D. Caird, M.J.M. Duke, Miniaturization of mechanical [78] A. Tuteja, W. Choi, J.M. Mabry, G.H. Mckinley, R.E. Cohen, Robust omniphobic
milling for powder X-ray diffraction, Powder Diffr. 27 (2012) 189–193, https:// surfaces, Proc. Natl. Acad. Sci. USA 105 (2008) 18200–18205. www.pnas.or
doi.org/10.1017/S0885715612000516. g/cgi/content/full/.
[53] G. Zhang, J.T. Germaine, R.T. Martin, A.J. Whittle, A simple sample-mounting [79] L. He, D. Wang, T. Ma, J. Song, Y. Wu, Y. Li, Y. Deng, G. Zhang, Processing and
method for radndom powder X-ray diffraction, Clay Clay Miner. 51 (2003) properties of a graphene-reinforced superhydrophobic siloxane, Mater. Des. 229
218–225, https://doi.org/10.1346/CCMN.2003.0510212. (2023), 111856, https://doi.org/10.1016/j.matdes.2023.111856.
[54] N. Doebelin, R. Kleeberg, Profex: a graphical user interface for the Rietveld [80] S. Diamond, Aspects of concrete porosity revisited, Cement Concr. Res. 29 (1999)
refinement program BGMN, J. Appl. Crystallogr. 48 (2015) 1573–1580, https:// 1181–1188.
doi.org/10.1107/S1600576715014685. [81] J.D. Birchall, A.J. Howard, K. Kendall, Flexural strength and porosity of cements,
[55] J. Ilavsky, P.R. Jemian, A.J. Allen, F. Zhang, L.E. Levine, G.G. Long, Ultra-small- Nature 289 (1981) 388–390.
angle X-ray scattering at the advanced photon source, J. Appl. Crystallogr. 42 [82] Y. Ma, J. Hu, G. Ye, The pore structure and permeability of alkali activated fly ash,
(2009) 469–479, https://doi.org/10.1107/S0021889809008802. Fuel 104 (2013) 771–780, https://doi.org/10.1016/j.fuel.2012.05.034.
[56] J. Ilavsky, F. Zhang, A.J. Allen, L.E. Levine, P.R. Jemian, G.G. Long, Ultra-small- [83] C.E. White, D.P. Olds, M. Hartl, R.P. Hjelm, K. Page, Evolution of the pore structure
angle X-ray scattering instrument at the advanced photon source: history, recent during the early stages of the alkali-activation reaction: an in situ small-angle
development, and current status, Metall. Mater. Trans. A Phys. Metall. Mater. Sci. neutron scattering investigation, J. Appl. Crystallogr. 50 (2017) 61–75, https://
44 (2013) 68–76, https://doi.org/10.1007/s11661-012-1431-y. doi.org/10.1107/S1600576716018331.
[57] J. Ilavsky, P.R. Jemian, Irena: tool suite for modeling and analysis of small-angle [84] R.F. Giese, P.M. Costanzo, C.J. van Oss, The surface free energies of talc and
scattering, J. Appl. Crystallogr. 42 (2009) 347–353, https://doi.org/10.1107/ pyrophyllite, Phys. Chem. Miner. 17 (1991) 611–616, https://doi.org/10.1007/
S0021889809002222. BF00203840.
[58] A.R. Hind, S.K. Bhargava, S.C. Grocott, The Surface Chemistry of Bayer Process [85] B. Pan, X. Yin, S. Iglauer, A review on clay wettability: from experimental
Solids: a Review, 1999. investigations to molecular dynamics simulations, Adv. Colloid Interface Sci. 285
[59] Q. Wan, F. Rao, S. Song, R.E. García, R.M. Estrella, C.L. Patiño, Y. Zhang, (2020), https://doi.org/10.1016/j.cis.2020.102266.
Geopolymerization reaction, microstructure and simulation of metakaolin-based [86] M.E. Schrader, S. Yariv, Wettability of clay minerals, J. Colloid Interface Sci. 136
geopolymers at extended Si/Al ratios, Cem. Concr. Compos. 79 (2017) 45–52, (1990) 85–94, https://doi.org/10.1016/0021-9797(90)90080-8.
https://doi.org/10.1016/j.cemconcomp.2017.01.014. [87] R. Šolc, M.H. Gerzabek, H. Lischka, D. Tunega, Wettability of kaolinite (001)
[60] J. Davidovits, Geopolymers, J. Therm. Anal. 37 (1991) 1633–1656, https://doi. surfaces - molecular dynamic study, Geoderma 169 (2011) 47–54, https://doi.org/
org/10.1007/BF01912193. 10.1016/j.geoderma.2011.02.004.
[61] M. Salahchini Javanmardi, E. Ameri, Pervaporation characteristics of PDMS/PMHS [88] J.T. Simpson, S.R. Hunter, T. Aytug, Superhydrophobic materials and coatings: a
nanocomposite membranes inclusive multi-walled carbon nanotubes for review, Rep. Prog. Phys. 78 (2015), 086501, https://doi.org/10.1088/0034-4885/
improvement of acetic acid–methanol esterification reaction, Polym. Bull. 77 78/8/086501.
(2020) 2591–2609, https://doi.org/10.1007/s00289-019-02875-z. [89] A. Das, J. Deka, K. Raidongia, U. Manna, Robust and self-healable bulk-
[62] P. Ferreira, Á. Carvalho, T.R. Correia, B.P. Antunes, I.J. Correia, P. Alves, superhydrophobic polymeric coating, Chem. Mater. 29 (2017) 8720–8728, https://
Functionalization of polydimethylsiloxane membranes to be used in the production doi.org/10.1021/ACS.CHEMMATER.7B02880/SUPPL_FILE/CM7B02880_SI_004.
of voice prostheses, Sci. Technol. Adv. Mater. 14 (2013), https://doi.org/10.1088/ AVI.
1468-6996/14/5/055006. [90] Z. Li, Z. Guo, Flexible 3D porous superhydrophobic composites for oil-water
[63] M. Izadifar, P. Thissen, A. Steudel, R. Kleeberg, S. Kaufhold, J. Kaltenbach, separation and organic solvent detection, Mater. Des. 196 (2020), https://doi.org/
R. Schuhmann, F. Dehn, K. Emmerich, Comprehensive examination of 10.1016/j.matdes.2020.109144.
dehydroxylation of kaolinite, disordered kaolinite, and dickite: experimental [91] C. Tournassat, A. Neaman, F. Villieras, D. Bosbach, L. Charlet, Nanomorphology of
studies and density functional theory, Clay Clay Miner. 68 (2020) 319–333, montmorillonite particles: estimation of the clay edge sorption sitedensity by low-
https://doi.org/10.1007/s42860-020-00082-w. pressure gas adsorption and AFM observations, Am. Mineral. 88 (2003)
[64] F. Toussaint, J.J. Fripiat, M.C. Gastuche, Dehydroxylation of kaolionite. I. Kinetics, 1989–1995.
J. Phys. Chem. 67 (1963) 26–30. [92] X. Liu, X. Lu, M. Sprik, J. Cheng, E.J. Meijer, R. Wang, Acidity of edge surface sites
[65] X. Zhu, Z. Zhu, X. Lei, C. Yan, Defects in structure as the sources of the surface of montmorillonite and kaolinite, Geochem. Cosmochim. Acta 117 (2013)
charges of kaolinite, Appl. Clay Sci. 124–125 (2016) 127–136, https://doi.org/ 180–190, https://doi.org/10.1016/j.gca.2013.04.008.
10.1016/j.clay.2016.01.033. [93] E.M. Pecini, M.J. Avena, Measuring the isoelectric point of the edges of clay
[66] I.C. Medeiros-Costa, E. Dib, N. Nesterenko, J.P. Dath, J.P. Gilson, S. Mintova, mineral particles: the case of montmorillonite, Langmuir 29 (2013) 14926–14934,
Silanol defect engineering and healing in zeolites: opportunities to fine-tune their https://doi.org/10.1021/la403384g.
properties and performances, Chem. Soc. Rev. 50 (2021) 11156–11179, https:// [94] R.N. Wenzel, Resistance of solid surfaces to wetting by water, Ind. Eng. Chem. 28
doi.org/10.1039/d1cs00395j. (1936) 988–994. https://pubs.acs.org/sharingguidelines.
15

Alkali Activated Organogeopolymers With Volumetri - 2024 - Cement and Concrete C

Uploaded by

Copyright:

Available Formats

Alkali Activated Organogeopolymers With Volumetri - 2024 - Cement and Concrete C

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Alkali Activated Organogeopolymers With Volumetri - 2024 - Cement and Concrete C

Uploaded by

Copyright:

Available Formats

Cement and Concrete Composites 145 (2024) 105336

Contents lists available at ScienceDirect

Cement and Concrete Composites

Alkali-activated organogeopolymers with volumetric superhydrophobicity

1. Introduction formula –(Si–O)z–Al(Na/K)–O– (where z = Si/Al molar ratio) have also

Data were provided by the manufacturers or suppliers.

2.6. Fourier transform infrared (FTIR) spectroscopy 3. Results and discussion

3.4. Incorporation of organic radicals

The heterofunctional co-condensation of the T units into the inor

of the sacrificed lower strength that may be improved by other tech

You might also like

Alkali Activated Organogeopolymers With Volumetri - 2024 - Cement and Concrete C

Uploaded by

Copyright:

Available Formats

Alkali Activated Organogeopolymers With Volumetri - 2024 - Cement and Concrete C

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Alkali Activated Organogeopolymers With Volumetri - 2024 - Cement and Concrete C

Uploaded by

Copyright:

Available Formats

Cement and Concrete Composites 145 (2024) 105336

Contents lists available at ScienceDirect

Cement and Concrete Composites

Alkali-activated organogeopolymers with volumetric superhydrophobicity

1. Introduction formula –(Si–O)z–Al(Na/K)–O– (where z = Si/Al molar ratio) have also

Data were provided by the manufacturers or suppliers.

2.6. Fourier transform infrared (FTIR) spectroscopy 3. Results and discussion

3.4. Incorporation of organic radicals

The heterofunctional co-condensation of the T units into the inor­

of the sacrificed lower strength that may be improved by other tech­

You might also like

The heterofunctional co-condensation of the T units into the inor

of the sacrificed lower strength that may be improved by other tech