Ceramics International xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Ceramics International
journal homepage: www.elsevier.com/locate/ceramint

Fabrication of silica aerogel composite blankets from an aqueous silica

aerogel slurry
⁎ ⁎
Kyoung-Jin Leea, Yeong-Ju Choea, Young Hun Kimb, Je Kyun Leeb, , Hae-Jin Hwanga,
a
Dep. Mater. Sci. & Eng., Inha University, 253 Yonghyun-dong, Nam-gu, Incheon, Republic of Korea
b
Basic Materials & Chemicals R & D, LG Chem R & D Campus Daejeon, 188 Munji-ro, Yuseong-gu, Daejeon, Republic of Korea

A R T I C L E I N F O A B S T R A C T

Keywords: Low-density silica aerogel composite blankets were successfully fabricated from silica aerogel slurries via an
Silica aerogels impregnation technique. Silica aerogel powder formed a stable slurry in a mixed ethanol/water solution with a
Blankets polyvinyl alcohol (PVA) binder. Scanning electron microscopy (SEM) images indicated that the silica aerogel
Porosity particles maintained their mesoporous and three-dimensional microstructure upon evaporation of the solvent.
Thermal conductivity
The obtained silica aerogel composite blankets had low densities (0.13 g cm−3), high silica aerogel contents
Polyvinyl alcohol (PVA)
(79%), and low thermal conductivities (18.92 mW m−1 K−1) compared to those of the polyethylene (PE)
blanket.

1. Introduction blankets manufactured by the above techniques is that silica aerogel

dust is released because the bonds between the silica aerogel particles
Silica (SiO2) aerogels are unique materials with ultra-low densities and fibers is based on very weak cohesion [11]. In addition, the drying
and highly cross-linked structures. The typical silica aerogel consists of requires long processing times, which increases the overall production
~ 95% air, i.e., pores, while the remaining 5% is composed of silica cost of the composite blanket.
nanoparticles that form a cross-linked three-dimensional network. In recent years, silica aerogel/polymer binder systems were pro-
Because of these structural characteristics, silica aerogels possess ex- posed in the literature, and the effect of the binder and processing
traordinary properties such as high specific surface areas, low thermal parameters, such as mixing methods, on the thermal conductivity of
conductivities, low dielectric constants, low acoustic velocities, and low silica aerogel composites was studied extensively [12–15]. The most
refractive indices [1–3]. important factors that should be considered in fabricating silica
Among those properties, the thermal conductivity of silica aerogel is aerogel/binder composite systems are ensuring homogeneous mixing of
as low as 0.020 W m−1 K−1, which is comparable to that of air the binder and controlling the structure of the pores. In particular, the
(0.025 W m−1·K−1). Thus, aerogels have received much attention as organic binder should not be incorporated inside the silica aerogel pore
next-generation thermal insulators. Ongoing research continues to structure or the thermal conductivity will increase compared to that of
focus on improving the insulation performance while lowering the the bare silica aerogel particles.
production costs of silica aerogels. Currently, the use of silica aerogels is Therefore, the purpose of this study is to develop silica aerogel
limited due to their complicated and expensive manufacturing process blankets with excellent thermal insulation performance at a low cost
[4]. For example, supercritically dried monolithic silica aerogels are compared to commercially-available silica aerogel blankets. First, we
very fragile and have low mechanical strength, which make them dif- prepared a stable ceramic slurry containing silica aerogel granules and
ficult to process for building applications [5]. In addition, granular a polyvinyl alcohol (PVA) or polyvinyl butyral (PVB) binder using a
aerogel-based insulation suffers from time-dependent setting problems mixed water/ethanol solution. Then, a silica aerogel blanket was fab-
that can lead to heat leakage through the void spaces [6]. ricated by impregnating a bare blanket in the silica aerogel slurry and
Fabricating aerogel blankets consisting of aerogel particles and or- evaporating the solvent under ambient conditions. The microstructure
ganic or inorganic fibers is one method to address these drawbacks. and thermal conductivity of the fabricated blanket were also in-
Generally, aerogel blankets have been fabricated by silica sol gel casting vestigated.
in porous fiber batting followed by drying under supercritical or am-
bient conditions [7–10]. To date, the main drawback of silica aerogel

⁎
Corresponding authors.
E-mail addresses: jekyun@lgchem.com (J.K. Lee), hjhwang@inha.ac.kr (H.-J. Hwang).

http://dx.doi.org/10.1016/j.ceramint.2017.10.176
Received 4 September 2017; Received in revised form 23 October 2017; Accepted 25 October 2017
0272-8842/ © 2017 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Please cite this article as: Lee, K.J., Ceramics International (2017), http://dx.doi.org/10.1016/j.ceramint.2017.10.176
K.-J. Lee et al. Ceramics International xxx (xxxx) xxx–xxx

2. Experimental procedure S-4200, Hitachi, Japan).

Silica aerogel slurries were prepared from commercially available

silica aerogel powder (Enova mt1100, Cabot Corp.), distilled water, 3. Results and discussion
ethanol (99.5%, Samchun Pure Chemical), and a water- or ethanol-
based binder solution. The average particle (granule) size of the silica Fig. 1 shows the dispersion stability of the silica aerogel slurries
aerogels was 10 µm. The binder solution contained 0.125 wt% PVA with different ethanol-to-water ratios. As seen in Fig. 1, the silica
(99%, Sigma Aldrich) or PVB (B-79, Solutia Inc.), and their molecular aerogel powder and the water/PVA solution was completely separated
weights were 89,000–98,000 and 50,000–80,000, respectively. in the slurry with an ethanol-to-water ratio of 0:10. As the lighter silica
The silica aerogel powder was dispersed in an ethanol/water solu- aerogel particles drifted upwards due to differences in density, a layer
tion with PVA or ethanol with PVB. The ethanol-to-water ratios used of the heavier aqueous PVA solution appeared at the bottom. This
were 10:0, 8:2, 6:4, 5:5, 4:6, and 0:10. Cyclopentane (98%, Sigma phenomenon is not surprising because the surface of silica aerogel
Aldrich) was added to the slurries as a foaming agent. Slurries with particles is super-hydrophobic, with a contact angle of 150° [17]. In
varying silica aerogel powder contents were prepared and investigated contrast, this was not the case in the slurry with an ethanol-to-water
for their physical properties and microstructure. Silica aerogel blankets ratio of 10:0. When this slurry was prepared, the silica aerogels seemed
were fabricated by impregnating polyethylene (PE) fiber blankets to form a well-deflocculated suspension. However, they tended to form
(Thinsulate Type G, 3 M) with the slurries and drying at 100 °C for 2 h agglomerates that settled relatively slowly in bulk, and the sediment
under ambient pressure condition. was relatively dense and less easily dispersed. In addition, the super-
0.025% (w/w) of silica aerogel slurries with different ethanol-to- natant solution in the alcohol system was cloudy.
water ratios were prepared and the aqueous slurries were placed in a Conversely, the stability of the slurries was significantly enhanced
Turbiscan™ LAB Stability Analyzer (Formulaction Co., France). The in the ethanol/water systems. Aerogel particles in slurries with ethanol-
stability analysis of the slurries was carried out as a variation of sta- to-water ratios of 8:2, 6:4, and 5:5 settled rapidly in the bulk. Moreover,
bility coefficient, TSI, Turbiscan Stability Index) with time [16]. The the supernatant solution above the dispersion is clear; the sediment was
density of the silica aerogel blankets was determined by their weight of relatively low packing density, and it was easily redispersed. In an
and volume. In addition, the thermal conductivity of the blankets was ethanol-to-water ratio of 4:6, most aerogel particles were separated and
measured using the heat flow metering method in accordance with agglomerated at the top, with a smaller number of aerogel particles
ASTM C518 and ISO DIS 8301 with a heat flow meter (HFM 436 dispersed at the bottom. The ratio of ethanol to water in the slurry
Lambda, NETZSCH, Germany). A silica aerogel blanket sample (25 cm required to achieve phase separation depends on the solid loading, i.e.,
× 25 cm) was placed between two flat plates, with the upper and lower silica aerogel content; as the solid loading increased, the ratio of
plates set at 35 °C and 15 °C, respectively. When thermal equilibrium ethanol to water at which the slurry stabilized increased. The stability
was reached, i.e., the temperatures of the upper and lower plates were of silica aerogel slurries was evaluated by Turbiscan and the variation
at a steady state, the thermal conductivity of the blanket could be es- in TSI value as a function of time was displayed in Fig. 1(c). For all
timated using Fourier's law. The thermal conductivity was calculated slurry samples, the TSI value increased as time was extended. The TSI
from the heat flux, the thickness of the blanket, and the temperature value of the aerogel slurry with ethanol-to-water ratio of 8:2 was much
gradient of two plates. Furthermore, the microstructure of the aerogels lower than those of slurries with ethanol-to-water ratio of 6:4, 5:5, and
was observed by field emission scanning electron microscopy (FESEM, 4:6. These results showed that the slurry with ethanol-to-water ratio of
8:2 was more than 6:4, 5:5, and 4:6 samples. Therefore, it can be

Fig. 1. Sedimentation behavior for aerogel slurries with different ethanol to water ratios: (a) the as-prepared slurries, (b) slurries after 24 h and (c) TSI value of aerogel slurries with
different ethanol-to-water ratio as a function of time.

2
K.-J. Lee et al. Ceramics International xxx (xxxx) xxx–xxx

Fig. 2. SEM images of aerogel powders. (a) The

starting silica aerogel powder. Aerogel powders
dried from slurries with ethanol-to-water ratios of (b)
6:4 with PVA binder and (c) 10:0 with PVB binder.

inferred that the optimum condition for the stable silica aerogel slurry not form a stable suspension or slurry when water used as the surface of
was 8: 2. the aerogel was completely modified by the hydrophobic CH3- func-
Fig. 2 shows SEM images of the silica aerogel powders. In particular, tional groups. The silica aerogel powders separated and finally floated
Fig. 2(a) shows the starting silica aerogel powder, and Fig. 2(b) and (c) to the top, as shown in Fig. 1.
show the aerogel powders dried from slurries with an ethanol-to-water For the ethanol-based sample, the aerogel powder did not maintain
ratio of 8:2 and PVA binder and with an ethanol to water ratio of 10:0 its stability in the slurry for prolonged times because of the relatively
and PVB binder, respectively. The SEM image (Fig. 2(a)) shows the large particle size (~ 10 µm) of the silica aerogel and lack of repulsive
typical silica aerogel microstructure formed, with a three-dimensional forces, which can reduce the agglomeration between silica particles.
network of nano-sized silica particles and meso-sized pores. There was Furthermore, PVB in the ethanol-based slurry seems to not act as a
no significant change in the microstructure of the silica aerogel powders dispersing agent in the slurry. In addition, it appears that the PVB
dried from the slurry with an ethanol-to-water ratio of 8:2 and PVA binder was incorporated inside the pores of the silica aerogel particles.
binder. However, the silica aerogel powder dried from the slurry with The increased particle size and reduced pore size observed in Fig. 2(c)
an ethanol-to-water ratio of 10:0 and PVB binder exhibited a different might be associated with the incorporation of the PVB binder into the
microstructure (Fig. 2(c)). The sizes of the silica particles increased to pore structure. If the PVB binder penetrated the silica aerogel pore
approximately 40–50 nm, and the network structure shrunk, causing structure, the thermal conductivity of the silica aerogel powder would
the meso-sized pores to disappear. significantly increase [18].
Fig. 3 is a schematic showing the different microstructures of two An interesting feature observed in this study is the high stability of
silica aerogel powders prepared from water, ethanol, and ethanol/ the silica aerogel particles in ethanol/water solutions. As the silica
water solution-based slurries. As expected, the silica aerogel powder did aerogel particles are hydrophobic, ethanol is speculated to

Fig. 3. Schematic showing the agglomeration or dispersion me-

chanism of silica aerogel particles in the water, ethanol, and
ethanol/water-based slurries.

3
K.-J. Lee et al. Ceramics International xxx (xxxx) xxx–xxx

Fig. 4. SEM images of the PE blanket/silica aerogel

composite with (a) PVA and (b) PVB.

preferentially adsorb onto the surface of the silica aerogel, and there-
fore, the pore in the aerogel particle was primarily filled with ethanol.
Furthermore, the solubility of PVA in ethanol is very limited, which
could prevent the PVA molecule from being incorporated into the silica
aerogel pore structure. As evident in Fig. 2(b), the ethanol/water so-
lution resulted in a stable slurry and finally the typical mesoporous
microstructure [12].
SEM images of the PE blanket/silica aerogel composites produced
via an impregnation method are shown in Fig. 4. In particular, Fig. 4(a)
and (b) represent composite blanket samples prepared from an ethanol-
to-water ratio of 8:2 and PVA and from an ethanol-to-water ratio of
10:0 and PVB. For the PE/silica aerogel composite with PVA binder, the
composite structure seemed to be homogeneous, and the silica aerogel
particles adhered relatively well to the PE blanket.
In addition, there was little detachment of the silica aerogel powder.
These observations suggest that PVA facilitated strong binding between
the silica aerogel powder and PE fibrous structure. In contrast, the PE/
silica aerogel composite blanket synthesized with PVB exhibited in-
homogeneity and significant detachment of the silica aerogel powders.
Fig. 5 shows the silica aerogel content and the bulk density of the
PE/silica aerogel composite blankets as a function of the solid loading
content. The silica aerogel content in the composite blanket increased
with increasing solid loading in the slurries with PVA binder. First, the
aerogel content in the composite blanket increased linearly with solid
loading until saturation at 5 wt% solid loading. The maximum aerogel
content was found to be approximately 80 wt% with respect to the PE
blanket because the space between PE fibers was limited. At 5 wt%
solid loading, all the pore space between PE fibers was filled with silica
aerogel powder.
In contrast, the bulk density of the composite blanket decreased as a
function of the silica solid loading content as shown in Fig. 5(b). This
phenomenon is somewhat reasonable and closely related to the silica
aerogel content in the composite blanket because the density of the
Fig. 5. (a) Silica aerogel content and (b) the bulk density of the PE/silica aerogel com-
silica aerogel powder was much lower than that of the PE blanket fi-
posite blanket with respect to the solid loading in the slurries.
bers. Although shrinkage, which is due to the PVA binder in the
blanket, occurred between the PE fibers, the bulk density of the com-
posite blanket was more heavily influenced by silica aerogel powders described. The thermal conductivity decreased from 25.25 to
with much lower densities than the PE fiber. Adding PVA binder to the 18.92 mW m−1 K−1 (or by 25%) for the PE blanket. In addition, in the
composite blanket resulted in shrinkage of the PE fibers that occurred case of PE nonwoven fabrics, the thermal conductivity decreased from
not only between PE fibers and aerogel powder but also between PE 35.79 to 30.64 mW m−1 K−1, a thermal conductivity reduction of al-
fibers. Thus, the bulk density of the composite blanket might increase as most 20%. This result suggests that the aerogel particles are effective in
aerogel powder was added to the PE blanket. However, the shrinkage lowering thermal conductivity and that the impregnation method for
effect that could lead to a decrease in the blanket volume was sup- aerogel blanket composites from the silica aerogel slurry is applicable
pressed by adding aerogel powders. Consequently, the bulk density of to various insulation materials. Commercially available silica aerogel
the aerogel blanket composite decreased with silica aerogel content. blankets have thermal conductivities of 18–20 mW m−1 K−1 and
Fig. 6 shows the SEM images of the silica aerogel composite blanket 23 mW m−1 K−1 [17,19]. So, the thermal conductivity (18.9 mW m-
prepared from the slurries with varying loading contents. As the solid 1 K-1) obtained in this study is low enough for commercial use in the
loading was increased, the silica aerogel powder was more densely field of superinsulation.
packed in the PE fiber blankets. Furthermore, 5 wt% solid loading in the
silica aerogel slurry appears to be adequate.
Table 1 gives the bulk density and thermal conductivity of the silica 4. Conclusions
aerogel composite blankets. The bare PE fiber blanket data is also
Silica aerogel slurries, in which ~ 10 µm of silica aerogel powders

4
K.-J. Lee et al. Ceramics International xxx (xxxx) xxx–xxx

Fig. 6. SEM images of the aerogel composite blan-

kets prepared from slurries with (a) 3 wt%, (b) 4 wt
%, (c) 5 wt%, and (d) 6 wt% solid loadings.

Table 1 [2] L. Kuchta, V. Fajnor, About the synthesis and thermal stability of SiO2-aerogel, J.
Therm. Anal. Calorim. 46 (1996) 515–520.
Density and thermal conductivities of PE blankets and blanket/aerogel composite.
[3] I. Smirnova, W. Arlt, Synthesis of silica aerogels: influence of the supercritical CO2
on the sol-gel process, J. Sol. Gel Sci. Technol. 28 (2003) 175–184.
Bare PE PE blanket/ Bare PE PE nonwoven
[4] R. Baetens, B.P. Jelle, A. Gustavsen, Aerogel insulation for building applications: a
blanket aerogel nonwoven fabrics/ state-of-the-art review, Energy Build. 43 (2011) 761–769.
composite fabrics aerogel [5] M. Moner-Girona, E. Martınez, A. Roig, J. Esteve, E. Molins, Mechanical properties
composite of silica aerogels measured by microindentation: influence of sol–gel processing
parameters and carbon addition, J. Non Cryst. Solids 285 (2001) 244–250.
Density (g cm−3) – 0.14 – 0.09 [6] A. Hoseini, C. McCague, M. Andisheh-Tadbir, M. Bahrami, Aerogel blankets: from
Thermal conductivity 25.25 18.92 35.79 30.64 mathematical modeling to material characterization and experimental analysis, Int.
(mW m−1 K−1) J. Heat. Mass Transf. 93 (2016) 1124–1131.
[7] B.E. Coffman, J.E. Fesmire, S. White, G. Gould, S. Augustynowicz, Aerogel blanket
insulation materials for cryogenic applications, 1218 (2010) pp. 913–92.0.
[8] R. Trifu, N. Bhobho, Flexible coherent insulating structures, US Pat., 2007173157,
were well-suspended in a PVA solution with an ethanol-to-water ratio 2007.
of 8:2, were successfully prepared, and silica aerogel composite blan- [9] S. Hong, I. Song, Y. Park, H. Yun, K. Hwang, Y. Rhee, The synthesis of silica aerogel
kets were fabricated by impregnating PE blankets with the silica aerogel in the macroporous ceramic structure by Sono-gel process and supercritical drying
process, J. Korean Ceram. Soc. 47 (2010) 553–559.
slurries. The silica aerogel powder was found to be stable in the [10] J. Ryu, Flexible aerogel superinsulation and its manufacture, Aspen Systems, Inc.,
ethanol/water solution, and the pore structure was not filled with PVA US Pat., 6068882, 2000.
after drying. In the process, the silica aerogel particles maintained their [11] M. Koebel, A. Rigacci, P. Achard, Aerogel-based thermal superinsulation: an over-
view, J. Sol. Gel Sci. Technol. 63 (2012) 315–339.
meso-porous and three-dimensional network structure. The obtained
[12] H.M. Kim, Y.J. Noh, J. Yu, S.Y. Kim, J.R. Youn, Silica aerogel/polyvinyl alcohol
silica aerogel composite blankets had 60–80 wt% of silica aerogel and (PVA) insulation composites with preserved aerogel pores using interfaces between
were light, with densities ranging from 0.14 to 0.20 g cm−3. The the superhydrophobic aerogel and hydrophilic PVA solution, Compos. Part A: Appl.
thermal conductivities of the silica aerogel blankets were 18.92 and Sci. Manuf. 75 (2015) 39–45.
[13] V.G. Parale, K. Lee, H. Park, Flexible and transparent silica aerogels: an overview, J.
30.64 mW m−1 K−1 for bare PE and PE nonwoven fabric, respectively. Korean Ceram. Soc. 54 (2017) 184–199.
These values were approximately 20% lower than those of bare PE and [14] P.L. Teh, M. Mariatti, A. Wagiman, K.S. Beh, Effect of curing agent on the properties
PE nonwoven fabrics without silica aerogel powders. of mineral silica filled epoxy composites, Polym. Compos. 29 (2008) 27–36.
[15] S.Y. Kim, Y.J. Noh, J. Lim, N. You, Silica aerogel/polyimide composites with pre-
served aerogel pores using multi-step curing, Macromol. Res. 22 (2014) 108–111.
Acknowledgement [16] S. Chibowski, M. Wiśniewska, T. Urban, Influence of solution pH on stability of
aluminum oxide suspension in presence of polyacrylic acid, Adsorption 16 (2010)
321–332.
This work was supported by LG Chem. [17] The Cabot Corporation Online, 〈http://www.cabotcorp.com/solutions/products-
plus/aerogel/particles〉, (Accessed 6 June 2017).
References [18] G.S. Kim, S.H. Hyun, Effect of mixing on thermal and mechanical properties of
aerogel-PVB composites, J. Mater. Sci. 38 (2003) 1961–1966.
[19] Aspen Aerogels Online, 〈http://www.aerogel.com/pyrogel〉, (Accessed 6 October
[1] J. Chandradass, S. Kang, D. Bae, Synthesis of silica aerogel blanket by ambient 2017).
drying method using water glass based precursor and glass wool modified by alu-
mina sol, J. Non Cryst. Solids 354 (2008) 4115–4119.