Construction and Building Materials 144 (2017) 35–41

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

High performance concrete incorporating ceramic waste powder as

large partial replacement of Portland cement
⇑
Dima M. Kannan a, Sherif H. Aboubakr b, Amr S. EL-Dieb a, Mahmoud M. Reda Taha b,
a
Civil and Environmental Engineering Department, United Arab Emirates University, United Arab Emirates
b
Department of Civil Engineering, University of New Mexico, USA

highlights

● High performance concrete mixtures incorporating up to 40% ceramic waste powder as large partial replacement of cement can be produced.
● High performance concrete mixture incorporating large content of ceramic waste powder showed high strength and excellent durability
performance.
● Ceramic waste powder acts primarily as a filler rather than pozzolanic material.
● Producing high performance concrete can be an excellent source for recylcing large quantities of ceramic waste powder.

articleinfo
abstract
Article history:
Received 22 October 2016 Ceramic waste powder (CWP) is produced during ceramic tile polishing with potential environmental
Received in revised form 27 February 2017 pollution. CWP is silica-rich, alumina-rich and fine particle size material. High performance concrete
Accepted 12 March 2017 (HPC) mixtures incorporating 10–40% CWP as replacement of Portland cement by mass were evaluated.
Mechanical, durability and microstructural investigations of HPC mixtures were performed. It is shown
that concrete incorporating CWP as large replacement of cement has high strength and excellent
Keywords: durabil- ity. Microstructure investigations showed that incorporating CWP did not make significant
Ceramic waste powder (CWP) difference in cement hydration compared with cement without CWP. Performance improvement is
High performance concrete explained by the low water/cement ratio of the reference mixture enabling CWP to create dense
Frattini test
packing particles.
© 2017 Published by Elsevier Ltd.

1. Introduction
a significant reduction of cement in HPC mixtures without detract-
ing from the concrete’s performance.
Portland cement production accounts for around 5% of the glo-
Irassar et al. [1] studied the utilization of CWP as pozzolanic
bal carbon dioxide (CO2) emissions, which is categorized as a major
materials and reported that incorporation of ceramic waste with
greenhouse gas. While increasing demand on concrete is essential
Portland cement simulates hydration due to enhancement of effec-
and associated with rapid global development and construction
tive water-to-cement ratio in the system. However, it was claimed
industry growth, cement is the primary and most expensive com-
that with replacement between 8 to 40% no pozzolan activity was
ponent of concrete mixture. Partial or full replacement of cement is
observed at early ages, while good pozzolanic activity was
considered a sustainable solution toward decreasing the environ-
observed at 28 days. Pokorny´ et al. [2] showed that
mental impact of cement production and will also contribute to
incorporation of CWP reduced compressive, bending strength, and
sustainable concrete. This paper investigates the possibility of uti-
thermal properties, while improving thermal insulation. For similar
lizing ceramic waste powder (CWP) as partial replacement of
ceramic waste, Vej- melková et al. [3] showed that CWP slowed
cement in high performance concrete (HPC) mixture that is charac-
compressive strength development, and the 28 days compressive
terized by relatively high cement content and low water content.
strength reaching 90% of the reference concrete with no CWP.
CWP produced during ceramic tile polishing contributes toward
Similarly, Heidari and Tava- koli [4] and Pacheco-Torgal and Jalali
environmental pollution. We suggest that using CWP can enable
[5] reported reduction in early age compressive strength of concrete
with an increase of CWP content but with minor strength reduction
at later ages. Wang and Tian [6] showed that pozzolanic activity
and strength of con- crete incorporating CWP preceded that
⇑ Corresponding author.
incorporating fly ash. How- ever, addition of CWP reduced the
E-mail address: mrtaha@unm.edu (M.M. Reda Taha). heat of hydration and increased

http://dx.doi.org/10.1016/j.conbuildmat.2017.03.115
0950-0618/© 2017 Published by Elsevier Ltd.
3 D.M. Kannan et al. / Construction and Building Materials 144 (2017) 35–41

shrinkage. The microstructure of mortars incorporating CWP

calcium oxide (CaO) at relatively small content of 3.53%, 3.19%,
demonstrated close-grained, dense, and reticular hydration gel.
1.89%, and 1.24%, respectively. The remaining 2.55% by mass
Reiterman et al. [7] reported that replacing 10% of cement by
repre- sents other relatively insignificant compounds, each with
CWP reduced the mechanical properties by 3%. On the other hand,
content less than 1%. Chemical analysis was performed using
Cheng et al. [8] reported that replacing cement by 10–40% ceramic
Rigaku ZSX Primus II wavelength dispersive X-ray fluorescence
waste resulted in reduction of concrete permeability. Cheng et al.
(XRF) spectrometer.
[9] also showed that carbonation resistance of concrete with
Type I Portland cement was used. The coarse aggregate (CA)
CWP is reduced, while sulphate corrosion resistance is
was natural crushed stone with nominal size of 19 mm, with a
enhanced. More- over, Mishra and Vasugi [10] showed that 67%
specific gravity of 2.65, and absorption of 1.0%. Two types of fine
of compressive strength of concrete incorporating CWP was
aggregate (FA) were used in the mixture namely crushed natural
achieved within 7 days, while 80% in 28 days, due to the
sand (FA, Crushed Stone) with a fineness modulus of 3.9 and speci-
effect of early hardening when producing concrete
fic gravity of 2.63, and dune sand (FA, Dune Sand) with a fineness
incorporating CWP with high silica con- tent and ground
modulus of 1.0 and specific gravity of 2.63. The two types of fine
granulated blast furnace slag. López et al. [11] showed that
aggregates were mixed to achieve a combined fineness modulus
substituting white CWP for sand resulted in slight improvement
of the fine aggregate ranging from 2.7 to 2.8. Mixing two types of
of compressive strength. Vijayalakshmi et al. [12] also suggested
sand was necessary as neither type solely meets ASTM standards
replacement of sand by granite CWP up to 15%. While
and thus cannot be used to produce concrete. A sand mixture ratio
researchers showed that concrete incorporating CWP showed
was determined to meet ASTM standards.
good mechanical and durability characteristics, a
The reference concrete mixture was designed to achieve a
microstructural investigation of concrete incorporating CWP is
slump value of 50 ± 10 mm. The measured slump of the reference
nec- essary to understand the role of CWP in cement hydration
mixture was 55 mm. Mixture proportions for the 50 MPa concrete
and con- crete strength development
grade (HPC-0, HPC-10, HPC-20, HPC-30, and HPC-40) with CWP
In this paper, Portland cement was partially replaced in the pro-
as
duction of HPC with CWP. An experimental program examining
cement replacement with different mass replacement of 0, 10, 20,
replacing 0, 10, 20, 30 and 40% of Portland cement in 50 MPa
30 and 40% of the Portland cement, respectively, are given in
HPC mixtures with CWP was developed. Mechanical and
Table 1.
durability characteristics of concrete were examined. Furthermore,
microstructural analysis of the cement paste (CP) mixtures incor-
porating CWP and pozzolanic reactivity tests of the CWP were 2.2. Fresh and hardened concrete testing
performed.
The effect of using CWP on fresh and hardened concrete proper-
ties was studied. For fresh concrete, slump, slump loss and setting
2. Experimental methods time were measured. For hardened concrete, compressive strength
development at age of 28 and 90 days were measured using
2.1. Materials 100 ×100 100× mm cubes. The rapid chloride permeability test
(RCPT) was conducted on concrete discs with a thickness of
The CWP was obtained from the final polishing process of cera- 50 mm cut from 100 mm diameter cylinders following ASTM
mic products from a ceramic factory in Abu Dhabi, United Arab C1202 [13]. Bulk electrical resistivity was measured using concrete
Emirates (UAE). The CWP was dried and ground to an average cylinders 100 × 200 mm. Chloride ion permeability and electrical
specific surface area 555 m2/kg. Scanning electron microscope resistivity were used as durability indices [14]. Permeable pores
(SEM) image of the ceramic powder showed that it consisted of were measured at 90 days of age using ASTM C642 [15].
irregular and angular particles that resemble the shape of cement
particles Fig. 1(a). Chemical analysis proved that CWP consists
2.3. Frattini test for pozzolanic activity
mainly of 69.4% SiO 2 and 18.2% Al2O3 by mass, as shown in Fig. 1
(b). The remaining 12.4% were other oxides including magnesium
Frattinit test [16] was performed to identify the pozzolanic
oxide (MgO), sodium oxide (Na 2O), potassium oxide (K2O), and
activity of CWP following BS EN 196-5:2011 [17]. In this test,

Fig. 1. CWP (a) SEM of CWP particles (b) Chemical analysis of CWP.
Table 1
Concrete mixture proportions of HPC with CWP replacement.

Mixture ingredients Mixture designation

HPC-0 HPC-10 HPC-20 HPC-30 HPC-40

Cement (kg/m3) 485 437 388 340 291
CWP (kg/m3) 0 48 97 145 194
Water (Liters) 208 208 208 208 208
FA, Dune Sand (kg/m3) 265 263 262 260 258
FA, Crushed Stone (kg/m 3) 397 395 392 390 388
CA, 19-mm max. (kg/m3) 993 988 981 975 968
w/cm* Ratio 0.43 0.43 0.43 0.43 0.43
*
w/cm: is water cementitious materials (cement + CWP) ratio.

20 g test samples (CP-0, CP-20, and CP-40) with CWP as cement

replacement with different weight replacement of 0, 20, and 40% The interpretation of the 29Si MAS-NMR spectrum provides identi-
of the Portland cement respectively, were prepared and mixed fication of the silica polymerization where the number of n sharing
with 100 ml of distilled water. The samples were then sealed in oxygen atoms is expressed as Qn where n is the number sharing
plastic bottles in an oven at 40 °C. After 8 and 28 days, the mixture oxygen atoms. Using the intensity fractions of Qns, the degree of
was cooled to ambient temperature and vacuum filtered through hydrations Dh of hydrated cement paste can be calculated as D h = -
I1 + I2 + I3 where In represents the corresponding intensity of Qn
2.7 lm nominal pore size filter paper (Whatman No. 542). The fil-
trate was analysed for OH— by titration against dilute connection respectively [18]. Finally, thermal gravimetric analysis
Hydrochloric (TGA)/differential scanning calorimetry (DSC) was performed to
acid (HCl) solution with methyl orange indicator. Also, filtrate investigate the chemical composition of cement hydration. The
solu- tion was analysed for Ca2+ by pH adjustment to 12.5 using TGA/DSC model SDT Q600 was used for this analysis and the test
sodium hydroxide solution (NaOH), followed by titration with was performed from 25 °C to 1000 °C, heated at 10 °C/min and
0.03 M dihy- drated disodium salt of ethylenediaminetetra- purged with nitrogen flow of 60 ml/min. The resulting recorded
acetic acid (EDTA) solution using Patton and Reeders indicator. resolution was 0.001 °C.
Prior standardization of the dilute HCl and EDTA solution using
sodium carbonate solu- tion and calcium carbonate with HCl
solution, respectively, was performed. Results lying on and
above the saturation line indicate no pozzolanic activity and
those lying below the saturation line indicate removal of Ca2+
from solution, which is attributed to poz- zolanic activity. It
should be noted that this procedure assumes no other source of
soluble calcium influenced the system [15].

2.4. Microstructural characterization

X-ray diffraction (XRD) measurements were conducted using

the Rigaku Smartlab system with a powder diffractometer using
a CuKa radiation and a wavelength, k = 0.154 nm to record the
XRD data. The operation of this system functioned under the open
attenuator of 40 kV and 40 mA with an optimization for scattering
angles, h, of 5° < h < 140°. Furthermore, 29Si Magic Angle Spinning
Nuclear Magnetic Resonance (29Si MAS-NMR) spectroscopy was
implemented using a 7 mm Bruker® rotor and cryomagnet (BZH
Fig. 3. Initial and final setting time of HPC with CWP replacement.
300/89) for all testing conducted. Solid-state NMR method was
performed with approximately 10,000 scans on each specimen.

Fig. 4. Compressive strength of HPC with CWP replacement.

Fig. 2. Slump and slump loss of HPC with CWP replacement.
 Fig. 5. Frattini test at (a) 8 and (b) 28 days of CP with CWP replacement.

3. Results and discussion

setting acceleration of 4, 3, and 46 min, respectively, compared
with the reference concrete mixture without CWP. This could be
The initial slump and slump values with time are shown in
a result of fine CWP particles acting as nucleation sites, which
Fig. 2. The control mixture achieved a slump value of 55 mm. As
result in uniform dispersion and improve hydration of cement par-
the replacement level of CWP increased, the initial slump value
ticles compared with cement without CWP. On the other hand, the
decreased except for the 20% and 30% replacement level. This
concrete mixture with 40% CWP showed 42 min of retardation in
could be attributed to the high specific surface area of the CWP
the final setting, as shown in Fig. 3. It should be highlighted that
compared with that of the cement, which is approximately more
than 1.5 times that of the cement. The slump values decreased with
time owing to the hydration of the cement. For the control
mixture, the slump was ‘‘zero” after 60 min. It was observed that
for the mixtures including CWP the slump loss was reduced (i.e.
slump retention increased), the slump values reached ‘‘zero” at
around 120 min (for HPC-20) indicating good slump retention.
This can be attributed to the fact that the CWP has very limited
immediate hydraulic reaction due to its very low CaO content. It is
also noted that mixture including 40% CWP showed slump loss
trend similar to that of concrete mixture without CWP. With all the
concrete mixtures being tested in the same laboratory temperature
and humidity conditions; this could be associated to the very high
sur- face area of CWP specifically at relatively high replacement
ratios. In the meantime, CWP resulted in limited acceleration in
concrete initial and final setting time, determined as per ASTM
C403 [19], up to 10% and 20% replacement levels but had
observable acceleration for 30%, while retardation was observed
for 40% replacement level. Mixtures with CWP of 10, 20, 30%
replacement levels showed final
Fig. 7. Electrical resistivity of HPC with CWP replacement.

Fig. 6. RCPT Results of HPC with CWP replacement. Fig. 8. Permeable pores of HPC with CWP replacement.
the addition of CWP increased the quantity of fine materials in the
mixtures so the content and rate of bleeding of these mixtures was
significantly reduced. With bleeding reduction and retarded set-
ting, plastic shrinkage cracking was observable especially for 40%
CWP.
Fig. 4 shows compressive strength development of HPC with
CWP replacement. At 28 days of age, the highest compressive
Fig. 9. XRD plot of cement paste (CP), CWP and CP with CWP replacement.
Fig. 10. 29Si MAS-NMR spectra of CWP and CP with CWP replacement.
strength value achieved was the control mixture (51.5 MPa). This
relatively high strength is due to the relatively high cement con-
tent in the mixture. As the amount of cement decreases gradually,
the 28 day compressive strength decreased by 15%, 17%, 18% and
20% for HPC-10, HPC-20, HPC-30, and HPC-40 respectively. After
90 days of age, the rate of compressive strength development
slightly increased for all concrete mixtures incorporating CWP.
All strength values were lower than that of the reference concrete
mixture by 10%, 18%, 17% and 15% for HPC-10, HPC-20, HPC-30,
and HPC-40 respectively. While the above results might suggest
that CWP might be acting as filler rather than a
binding/pozzolanic material, Frattini test to examine pozzolanic
activity showed a dif- ferent indicator. Farttini test showed that
concrete with 20% and 40% CWP replacement of Portland cement
exhibited pozzolanic activity at 8 and 28 days age of concrete as
shown in Fig. 5. The decrease in compressive strength might be
attributed to the reduc- tion in the cement binder content and the
very low slump, which will apparently affect its comapactibility. It
is apparent that CWP will provide a relatively high silica
environment that might be able to convert Calcium hydroxide (CH)
to strong Calcium silicate hydrate C-S-H.
The results of chloride ion permeability measured in terms of
the electrical charge passing through the concrete discs in cou-
lombs are shown in Fig. 6. The permeability of chloride ions
decreased as the amount of CWP increased in all mixtures. This
trend was also similar within the individual mixtures as the age
 Fig. 11. (a) TGA and (b) DSC curves of CWP and CP with CWP replacement.

increased from 28 to 90 days. For 28, days RCPT decreased by 38%,

incorporating CWP is superior to that of concrete without CWP.
63%, 90%, and 89% for HPC-10, HPC-20, HPC-30, and HPC-40,
The chloride ion permeability significantly decreased and electrical
respectively, from that of the control mixture. Moreover, electrical
resistivity of concrete was significantly improved. No change in the
resistivity increased as the amount of CWP increased in all mix-
permeable pores was observed by incorporating CWP. Finally,
tures by 12%, 272%, 389%, and 516% at 28 days, 77%, 297%, 402%,
microstructural characterization showed that both 20% and 40%
and 665% at 90 days for HPC-10, HPC-20, HPC-30, and HPC-40,
replacement of cement with CWP resulted in similar 29Si MAS-
respectively, from that of the control mixture as shown in Fig. 7.
NMR peaks as that of hydrated Portland cement. Moreover, XRD
It is evident that incorporating CWP has significantly improved
and TGA/DSC confirms that there was no significant difference in
the durability characteristics of concrete creating HPC. This can
hydration components as a result of mixtures incorporating CWP.
be attributed to the fine particle size distribution of CWP compared
Nevertheless, the Frattini test showed CWP to have an observable
with cement, therefore, incorporating CWP might enable creating a
pozzolanic reactivity. It is evident CWP clearly has two effects;
dense packing particle system always reported as a key need for
first, a filler effect enabling it to create a dense packing system that
high and ultra-high performance concrete [20,21]. Fig. 8 shows
favours high performance concrete. Second, a pozzolanic effect
the effect of using CWP on permeable pores. It is noted that the
that is more pronounced at later ages compared with early age.
inclusion of CWP resulted in reducing permeable pores by 3% up
However, a sufficient amount of CH shall be available in the
to 24% compared with concrete mixture without CWP.
hydrated concrete mixture for the pozzolanic activity of CWP to
Microstructural investigations confirmed the insufficient
take place. The limited pozzolanic reactivity in the examined HPC
amounts of CH to react with all the available silica when CWP is
mixes can be attributed to the lack of water in the concrete mix-
incorporated. XRD spectrographs show that Crystalline SiO 2, C-S-
ture given the relatively low water/cement ratio in the reference
H [22], CH species detected with no significant difference for
concrete mixture.
CWP replacement ratio of 20% and 40% as shown in Fig. 9. Simi-
larly, 29Si MAS-NMR results show that CP-0 is characterized by
Q0, Q1, Q2 silica polymerization, while CWP is characterized by
Acknowledgments
Q3, Q4 silica polymerization. For 20% and 40% replacement, ceramic
cement paste peaks of both CP-0 and CWP were detected as shown
This work was financially supported by the UAEU-UPAR2
in Fig. 10. The degree of hydration was found to be 61%, 64% and
Research Grant # 31N208. Also, the donation of the ceramic waste
65.7% for cement, 20% and 40% replacement respectively. This dif-
powder for the study by PORCELLAN (ICAD II MUSSAFAH –
ference is insignificant and within experimental error proving the
ABU DHABI, UAE) and the cooperation of Eng. Mostafa Gad Alla
fact that CWP acts as a filling material. Moreover, no significant
and Mr. Dilip Kumar Borah are highly appreciated. Support to the
dif- ference was observed between TGA and DSC graphs as shown
sec- ond and last authors by Southern Plains Transportation Centre
in Fig. 11. These conclusions confirm there is that no significant dif-
(SPTC) to University of New Mexico is greatly appreciated.
ference in the hydration components in mixtures incorporating
CWP and the reference concrete mixture. Nevertheless, CWP was
still able to produce high performance concrete with 50 MPa
References
strength at 90 days, with very low chloride ion permeability and
significantly very high electrical resistivity. [1] E. Irassar, V. Rahhal, A. Tironi, M. Trezza, Z. Pavlík, M. Pavlíková, M. Jerman, R. Cerny´
, Utilization of ceramic wastes as pozzolanic materials, in: Technical Proc.
of the 2014 NSTI Nanotechnology Conference and Expo, NSTI-Nanotech 2014, vol.
4. Conclusions 3, 2014, pp. 230–233.
[2] J. Pokorny´ , J. Forˇt, M. Pavlíková, J. Studnika, Z. Pavlík, Application of
mixed ceramic powder in cement based composites, Adv. Mater. Res. 1054
The above work suggests that high performance concrete can be (2014) 177–181.
produced with significant replacement between 20 and 40% of [3] E. Vejmelková, D. Koáková, T. Kulovaná, A. Hubáek, R. Erny´ , Mechanical
and thermal properties of moderate-strength concrete with ceramic powder used
Portland cement with CWP. The resulting concrete will have a
as supplementary cementitious material, Adv. Mater. Res. 1054 (2014) 194– 198.
lower compressive strength at early age and comparable compres- [4] A. Heidari, D. Tavakoli, A study of the mechanical properties of ground
sive strength at late ages compared with reference normal concrete ceramic powder concrete incorporating nano-SiO2 particles, Const. Bldg. Mater. 38 (2013) 255–
mixtures without CWP. However, durability of concrete mixtures 264.
 [5] F. Pacheco-Torgal, S. Jalali, Compressive strength and durability properties of ceramic L. Basheer, L. Kropp, D.J. Cleland, Assessment of the durability of concrete from its
wastes based concrete, Mater. Struct./Materiaux et Constructions 44 (1) (2011) 155– permeation properties: a review, Const. Bldg. Mater. 15 (2–3) (2001) 93– 103.
167.
[6] G. Wang, B. Tian, Effect of waste ceramic polishing powder on the properties of
cement mortars, International Conf. on Energy and Environ. Tech., ICEET 2009,
vol. 1, 2009, pp. 101–104.
[7] P. Reiterman, O. Holcapek, M. Cachova, F. Vogel, M. Jogl, P. Konvalinka, Basic and
hygric properties of concrete containing fine ceramic powder, Adv. Mater. Res. 897
(2014) 188–191.
[8] Y. Cheng, F. Huang, R. Liu, J.-L. Hou, G.-L. Li, Test research on effects of waste
ceramic polishing powder on the permeability resistance of concrete, Mater.
Struct. 9 (3) (2016) 729–738.
[9] Y. Cheng, F. Huang, G.-L. Li, L. Xu, J.-L Hou, Test research on effects of ceramic
polishing powder on carbonation and sulphate-corrosion resistance of concrete, Const.
Bldg. Mater. 55 (2014) 440–446.
[10] N. Mishra, V. Vasugi, Experimental assessment of properties of ternary
blended concrete using GGBS and ceramic powder, Indian Conc. J. 89 (7) (2015)
74–80.
[11] V. López, B. Llamas, A. Juan, J.M. Morán, I. Guerra, Eco-efficient concretes:
impact of the use of white ceramic powder on the mechanical properties of concrete,
Biosys. Eng. 96 (4) (2007) 559–564.
[12] M. Vijayalakshmi, A.S.S. Sekar, M. Sivabharathy, G. Ganesh Prabhu, Utilization
of granite powder waste in concrete production, Defect Diffusion Forum 330 (2012)
49–61.
[13] ASTM C1202-12, Standard Test Method for Electrical Indication of Concrete’s
Ability to Resist Chloride Ion Penetration, ASTM International, West
Conshohocken, Pennsylvania, 2012.
[14]
[15] ASTM C642-13, Standard Test Method for Density, Absorption and Voids in
Hardened Concrete, ASTM International, West Conshohocken, Pennsylvania,
2012.
[16] S. Donatello, M. Tyrer, C.R. Cheeseman, Comparison of test methods to assess
pozzolanic activity, Cem. Conc. Comp. 32 (2) (2010) 121–127.
[17] BS EN 196-5:2011, Methods of testing cement - Part 5: Pozzolanicity test for
pozzolanic cement, British Standards Institution, London, UK, 2011.
[18] G.M.M. Bell, J. Bensted, F.P. Glasser, E.E. Lachowski, D.R. Roberts, M.J.
Taylor, Study of calcium silicate hydrates by solid state high resolution 29Si nuclear magnetic resonance,
Adv. Cem. Res. 3 (9) (1990) 23–27.
[19] ASTM C403/C403M-16, Standard Test Method for Time of Setting of Concrete
Mixtures by Penetration Resistance, ASTM International, West Conshohocken,
Pennsylvania, 2016.
[20] P.K. Mehta, P.C. Aitcin, Microstructural basis of selection of materials and mix
proportions for high-strength concrete, high strength concrete, in: W.T. Hester
(Ed.), 2nd Int. Symp., American Concrete Institute SP-121, Detroit, 1990, pp.
265–286.
[21] M.M. Reda, N.G. Shrive, J.E. Gillott, Microstructural investigations of ultra high
performance concrete, Cem. Concr. Res. 29 (1999) 323–329.
[22] S. Grangeon, F. Claret, Y. Linard, C. Chiaberge, X-ray diffraction: a powerful tool
to probe and understand the structure of nanocrystalline calcium silicate
hydrates, Acta Cryst. B 69 (5) (2013). 465-286.