1 Introduction

Highly porous materials (HPMs) possess many practically useful properties. Polymeric HPMs exhibit good acoustic- and vibration-insulating features, while ceramic materials have excellent thermal insulation and fireproof characteristics. Therefore heat-shielding materials based on silica fibers (for example, TZMK, Russia; Li-900 and Li-2200, USA) are used in aerospace technology [1,2,3,4,5]. The HPM properties mainly depend on the chemical composition, although topological structure—the pore form (open, closed), size (ranging from nanometers to hundreds of microns) is also important [1, 6].

Nowadays there is a need to expand the range of materials with new functional and operational characteristics. The creation of new complex hierarchical highly-porous materials, where HPMs with smaller pores are introduced into large pores of another HPMs is of considerable interest [7]. The multimodal combination of components is attracting increased attention due to the fact that the combination of high strength of macroporous materials with an extremely developed surface of the nanoporous component can provide improved performance characteristics of the hybrid material.

One of the limits of the ceramic HPMs’ use is their high hydrophilicity. They absorb large amount of water with deterioration of heat-protective properties of the material, and sometimes leading to its destruction. For this reason, the search of new hydrophobization methods of such materials is needed. Silica-based materials are usually made hydrophobic via treatment by trimethylchlorosilane (TMCS), hexamethyldisilazane (HMDZ) and related compounds [8,9,10,11], but we could not find data on hydrophobization of HPMs by this method. The standard method for increasing the HPMs’ hydrophobicity is impregnation with polydimethylsiloxane, and it is shown that, depending on the processing temperature, siloxane is either not bonded with the surface or the formation of sufficiently strong covalent bonds occurs [12]. It is commonly known that fluoroolygomers can be used to provide hydrophobicity for materials’ surface [13, 14]. The dissolution of fluoroparaffins in supercritical carbon dioxide (SC–CO2), followed by precipitation on the surface can be used for hydrophobization of HPMs fibers [15]. Each technique has advantages and limitations, and therefore it is useful to search for new technological methods of HPMs’ hydrophobization.

In the present work, the increase in the hydrophobicity of a ceramic porous TZMK-material was performed by filling the micron-size pores with a hydrophobic methyltrimethoxysilane-based aerogel (MTMS, CH3Si(OCH3)3) [16,17,18,19,20].

Aerogels are solid mesoporous materials with a typical particle size in the range of 2–20 nm. Aerogels are characterized by a number of unique properties such as extremely low thermal conductivity, very low density, high specific surface area, high porosity, high acoustic-absorbing ability, often—transparency. Aerogels and aerogel-based materials can be used as ultralight construction materials or additives for the modification of existing construction materials, effective desiccants and sorbents, heterogeneous catalysts and catalyst carriers, as thermal- and acoustic-insulating materials, biomaterials, etc. [21,22,23,24].

In this paper, we have developed a method of the preparation of a hydrophobic composite material with bimodal porosity based on a highly porous ceramic material (TZMK) with micron-sized pores and a hydrophobic aerogel with nanosized pores. The alternative inexpensive method of SiO2 surface hydrophobization—treatment by TMCS—was also applied for comparison purpose.

2 Materials and methods

2.1 Materials

Methyltrimethoxysilane (97%), trimethylchlorosilane (98%) isopropanol (IP, 99.5+%), HF (40% aqueous solution), and ammonium carbonate were used as received. All chemicals were purchased from Acros.

2.2 Preparation of TZMK, gel and composite gel material

High-temperature heat-shielding and thermal-insulating materials (TZMK) are made of refractory oxide fibers elongated from a natural quartz melt with a purity of 99.9%. Samples of highly porous fibrous material with a size of 10 × 10 × 5 mm were made by vacuum formation of an aqueous fiber pulp at a weight ratio of 15:1 with the addition of 1 mass % of amorphous boron [1].

Composite preparation was carried out according to a three-stage scheme: (1) impregnation of the TZMK sample with an excess of colloidal MTMS-based solution. (2) Gelation of sol in TZMK macropores. (3) Supercritical drying using isopropyl alcohol or CO2 as a supercritical fluid. Individual MTMS-based aerogels were prepared by supercritical drying only in isopropyl alcohol.

The MTMS-based sol was prepared according to the procedure described in the literature [19]. 0.955 g (0.007 mol) of MTMS was dissolved in 1.26 g (0.021 mol) of isopropanol, then 40% aqueous HF (0.025 g, 0.0005 mol) in 0.5 g (0.028 mol) of water was added, and the mixture was stirred for 1 h. Afterwards, 0.064 g (0.00067 mol) of ammonium carbonate dissolved in 0.6 g (0.034 mol) of water was added and stirred for 1 min. The samples of TZMK were filled with obtained sol by soaking, placed in polypropylene containers, gelation occurred within 30–40 min. The containers were then kept at room temperature for 24 h, after which the samples were washed with isopropanol once a day for 5 days.

2.3 Supercritical drying

Supercritical drying was performed as follows: a composite gel sample in a glass tube containing 14–16 mL of isopropanol was placed into a stainless steel autoclave (V ~ 40 mL). The autoclave was sealed and heated to a temperature 250–260 °C (the measured pressure in the autoclave at the beginning of the drying procedure reached 6.0–7.0 MPa) exceeding the critical temperature of the isopropanol (235 °C). The heating rate was approximately 100 °C/h. After reaching the desired temperature, the valve was opened; the pressure was evenly decreased to atmospheric over 2 h. The hot autoclave was then evacuated for 30 min, cooled to room temperature and opened. Excessive aerogel was removed by a soft brush from the composite surface.

Supercritical drying in CO2 was carried out in an installation composed of a high pressure CO2 pump (SSI Supercritical 24, USA), a 50 mL steel reactor, and a Waters BPR (USA) back pressure regulator. The isopropanol-washed composite sample was washed with liquid CO2 for 2 h at 20 °C at a pressure of 15 MPa, then the temperature in the reactor was elevated to 50 °C, and the sample was washed with supercritical CO2 (15 MPa) for 2–2.5 h. Next, the pressure in the heated autoclave was gradually decreased to atmospheric; the autoclave was cooled to ambient temperature and opened.

Silylation of TZMK by TMCS was performed as follows. 2.5 g of TZMK, 15 mL of hexane and 1 mL of TMCS were placed into a glass flask equipped with a reflux condenser. The mixture was heated with reflux for 5 h. Silylated sample was then washed by hexane three times and dried at ambient temperature and pressure.

2.4 Characterization of materials obtained

Characterization of materials was performed using the equipment of the IGIC RAS Joint Research Centre for Physical Methods of Research.

The microstructure of the samples was studied using Hitachi SU 8010 and Carl ZeissNVision 40 high-resolution scanning electron microscopes (SEM) at 1 kV acceleration voltage.

The specific surface area and porosity of aerogels were determined by low-temperature nitrogen adsorption measurements with Katakon ATX-06 analyzers. The specific surface area values were determined using a 5-point Brunauer, Emmett and Teller (BET) method at the relative pressure range of P/P0 = 0.05–0.25; the correlation coefficients of the corresponding linear regressions in the coordinates of the BET equation were not less than 0.9975. Pore volume distributions were determined according to the Barrett, Joyner and Halenda (BJH) method, by the analysis of a desorption branch of an adsorption-desorption isotherm (28 points). Prior to measurements, samples were flushed with a dry nitrogen flow for 30 min at 200 °C.

Sessile water drop contact angles were measured on an automatic optical analyzer ОСА 15 Pro DataPhysics (± 0.1° accuracy, at least twelve measurements for each sample). The volume of water drops was fixed at ~ 30 µL. To measure water absorption the samples were immersed into water and kept under surface at a temperature of 22 ± 5 °C, then they were weighed on the analytical balance АВ-210 with an accuracy of 0.1 mg.

Determination of the compressive strength before reaching 10% deformation of the samples (the selection of 8 samples) was carried out on the Instron 5965 unit with a traverse speed of 2 mm/min perpendicular to the preferred direction of fiber placement in the sample.

Thermogravimetric analysis was conducted with the use of a Netzsch STA 449 C Jupiter analyzer. The analysis was performed in air at heating rate 10 K/min up to 800 °C (± 1 K accuracy).

X-ray powder diffraction (XRD) patterns were recorded with a Bruker D8 Advance diffractometer, using CuKα radiation in the 2θ range 10°–90° at a 2θ step of 0.02° and a counting time of 0.3 s per step.

3 Results and discussion

In this paper we investigated samples supercritically dried in isopropyl alcohol (denoted as AG@TZMK-IP) and in CO2 (AG@TZMK-CO2). TZMK treated by trimethylchlorosilane is denoted as TZMK-TMCS.

It can be seen from the micrograph (Fig. 1) that pristine TZMK consists of long fibers of a few µm diameters. Fibers are fastened in places of contact, providing rigidity and macroscopic shape of the sample. The pore size varies from 10 to 30 μm.

Fig. 1
figure 1

SEM image of TZMK

Full size image

The SEM images (Fig. 2) presents the structure of the obtained hydrophobic composite materials.

Fig. 2
figure 2

SEM images of obtained hydrophobic composite materials AG@TZMK-IP (a and c) and AG@TZMK-CO2 (b)

Full size image

It can be seen, that TZMK pores are filled with spherical MTMS-based aerogel particles. In addition, TZMK fibers have an almost continuous coating with the MTMS layer (Fig. 3).

Fig. 3
figure 3

SEM images of TZMK fiber of AG@TZMK-IP

Full size image

As shown previously [25], treatment of hydrophilic SiO2 aerogels with methyltrialkoxysilane leads to the production of highly hydrophobic materials (a contact angle value of about 150°) due to the formation of a layer of Si–CH3 groups on the SiO2 surface (Scheme 1). It is reasonable to assume that the hydroxylated surface of the SiO2 TZMK fiber is modified according to a similar scheme.

Scheme 1
scheme 1

Binding of hydrophobic groups of methyltrimethoxysilane with the surface of SiO2 fibers

Full size image

A hydrophobic layer of Si–CH3-groups is bound to the surface by covalent Si–O–Si bonds and does not break down when treated with water or mechanical action. This supposition is confirmed by the fact that the surface of SiO2 was earlier coated with silanes containing different functional groups, for example, Si–CH2CH2CH2–NH–CH2CH2–NH2 and Si–CH2CH2CH2–SH [26].

There are probably two parallel processes in our case: (a) hydrophobization of SiO2 surface according to a process on Scheme 1 and (b) the formation of MTMS-based gel particles filling the TZMK pores and preventing water penetration into the material.

These processes apparently impart hydrophobic properties to the fibers of the investigated composite material, both by replacing the polar silanol groups with nonpolar Si–O–Si–CH3 groups and by providing a roughness of the smooth surface of the fibers. It is known that the hydrophobicity of the surface and/or the creation of a polymodal relief on it are the conditions providing the high hydrophobicity of the material [27, 28].

The water drop contact angle measurements of composites and TMCS-treated TZMK sample showed a high degree of hydrophobicity: AG@TZMK-IP being the most hydrophobic (Table 1; Fig. 4).

Fig. 4
figure 4

The photograph of a sessile water drop on the surface of a hydrophobic AG@TZMK-IP

Full size image
Table 1 Contact angle measurments of TZMK-materials
Full size table

Only a slight decrease in the contact angle to a value of 142° is observed after exposure of the AG@TZMK-IP under 100% humidity for 14 days at a temperature of 20 °C. The decrease can be caused by the hydrolytic cleavage of Si–O–Si bonds at continuous contact with water leading to the formation of silanol groups [29]. Nevertheless, this makes it possible to assert the stability of the highly hydrophobic properties of the composite and the possibility of its use under conditions of high humidity for a long time. AG@TZMK-IP floats on the water surface during more than a year (Fig. 5), whereas the pristine TZMK sample sinks after several seconds.

Fig. 5
figure 5

AG@TZMK-IP floating on the water surface

Full size image

Figure 6 presents the water absorption histogram of MTMS-based aerogel composite samples and TZMK treated with trimethylchlorosilane after exposure under water surface. AG@TZMK-IP exhibits an order of magnitude greater resistance to water absorption: the mass of absorbed water after 480 h of exposure is 5.9% by weight, whereas AG@TZMK-CO2–76.9% and TZMK-TMCS—50%. This fact can be explained by the strong covalent grafting of hydrophobic aerogel particles to the surface of oxide fibers due to a higher drying temperature in supercritical isopropanol (250 °C) compared with 50 °C in SC–CO2. For comparison the water absorption of the initial TZMK sample is more than 500% by weight after the first day of exposure. A relatively high water absorption by the TZMK-TMCS sample is not clear yet. Probably, the places of contact between SiO2 rods are filled with lower-melting SiO2–B2O3 eutectic serving as an adhesive and are not covered by trimethylsilyl groups giving the way to water molecules.

Fig. 6
figure 6

The water absorption histogram of TZMK-materials

Full size image

Changes in hydrophobicity of AG samples cannot be explained by recrystallization of TZMK matrix during hydrophobization procedure. According to XRD pristine TZMK, TZMK-TMCS and AG@TZMK-CO2 samples consist from amorphous fused silica. XRD patterns for TZMK and AG@TZMK-IP samples are presented in Supplementary (Fig. S1).

AG@TZMK-IP and TZMK-TMCS materials were subjected to additional tests.

We have shown previously that a mechanical failure (the appearance of cracks and chips) occurs after two cycles of thermal impact shock stress (heating to + 25 °C and subsequent cooling to – 18 °C) of the water-saturated TZMK-sample (Fig. 7) [30].

Fig. 7
figure 7

The appearance of water-saturated TZMK-sample after two cycles of thermal stresses

Full size image

In contrast AG@TZMK-IP sample (Fig. 8a) is not destroyed after a threefold cyclic freeze-thaw action, owing to lower moisture content in comparison to the TZMK-sample (the final water content in AG@TZMK-IP was only 0.27%). It is worth mentioning that TZMK-TMCS sample sank in water after the first freeze-thaw cycle and sample destruction was observed (Fig. 8b). Water content in this sample was approximately 370%.

Fig. 8
figure 8

AG@TZMK-IP (a) and TZMK-TMCS (b) after a threefold cyclic freeze-thaw action

Full size image

The results of some additional characterization of AG@TZMK-IP and TZMK-TMCS are presented in Table 2.

Table 2 The characteristics of highly porous materials
Full size table

The mechanical strength is an important factor determining the practical use of the HPMs. Therefore, the density and compressive strength to 10% deformation of TZMK, MTMS-based aerogel and AG@TZMK-IP samples were determined (Table 2).

It should be noted that the value of compressive strength of the AG@TZMK-IP composite is more than two times higher than that of pristine TZMK – 1.14 vs. 0.55 MPa—demonstrating a strong synergism of the mechanical strength even when adding a very soft aerogel. Such a kind of synergism with the participation of MTMS-based aerogel is described earlier [31]. A significant increase in strength is important for the practical use of the material. The resiliency of the composite is determined by brittle TZMK and is significantly lower than that of MTMS-based composites with aramid [32] and pectin [33].

The increase in composite density in comparison with TZMK is obviously due to the filling of the pores with the MTMS-based aerogel and with the deposition of thin coatings on oxide fibers.

The thermal conductivity of the composite is twice as high as that of pristine TZMK. This can be explained by the enhanced solid backbone thermal conductivity of the composite due to its higher density. The thermal conductivity of fused silica (~ 2–4 W/K m) is higher than that of TZMK itself. We couldn’t find any data on the thermal conductivity of MTMS-based backbone (which can chemically be presented as CH3SiO1.5) but if to assume that it is close to fused silica it becomes clear that the thermal conductivity of the composite must increase.

The presence of a mesoporous MTMS-based aerogel in the AG@TZMK-IP could significantly change the specific surface area of the sample as compared with the parent TZMK. Indeed the studies did show an increase in the specific surface area by more than three orders of magnitude (from 0.12 for TZMK to 195 m2/g for the composite).

Figure 9 shows the full adsorption/desorption isotherm as well as the pore size distribution of AG@TZMK-IP, the volume of nanopores being 0.041 mL/g. The isotherm is of type IV and is characterized by a narrow and are characterized by a narrow hysteresis of type H3, typical for materials containing slit-like pores [20]. The pores are mainly smaller than 10 nm, but larger pores also present.

Fig. 9
figure 9

Full adsorption/desorption isotherm and corresponding pore size distribution of AG@TZMK-IP

Full size image

To determine the temperature behavior of the composite material we studied the thermal properties of AG@TZMK-IP. The original TZMK is stable in an oxidizing atmosphere at temperatures above 1000 °C and does not limit the thermal stability range of the composite. The differential thermal analysis and thermogravimetric analysis of the pure MTMS-based aerogel (dried in isopropanol) showed its relatively high thermal stability (Fig. 10a). The weight loss begins at 180 °C, the main stage of thermal decomposition was observed between 370 and 460 °C, and the weight of the sample became constant at 560 °C. The total weight loss observed was about 13%, which is obviously connected with the oxidation of CH3-groups in MTMS-based aerogels by air oxygen. MTMS-based aerogel formula can be nominally represented as CH3–SiO1.5, thus in the case of its complete oxidation and formation of SiO2 the theoretical weight loss is equal to 10.4%. Higher experimental value for weight loss of MTMS-based aerogel is probably connected to the excess carbon content in the material which may be due to the surface modification of aerogel matrix with alkyl moieties during supercritical drying stage (Si–OH + R–OH → Si–OR). Such a modification is commonly observed in silica aerogels synthesized via high-temperature supercritical drying [33,34,35,36].

Fig. 10
figure 10

The differential thermal analysis and thermogravimetric analysis of the MTMS-aerogel (a) and AG@TZMK-IP (b)

Full size image

The thermal behaviour of the composite material (Fig. 10b) is more complex. The main exothermic peaks most probably correspond to the oxidation of the CH-bonds in the isopropyl-groups (395 °C) which are present in aerogel after drying in isopropyl alcohol [33], and of silicon-bonded methyl group of the aerogel matrix (500 °C). The total weight loss of the composite upon heating up to 700 °C is ~ 5%. It should be noted that the heat resistance of the composite significantly exceeds the known hydrophobic HPMs. For example, TZMK composite obtained by dissolving of fluoroparaffins in supercritical carbon dioxide withstands temperatures not higher than 200 °C [15].

4 Conclusions

A multi-porous composite material formed by a ceramic fiber TZMK and MTMS-based aerogel was obtained for the first time. Carbon dioxide and isopropanol were used as supercritical solvents and it was shown that the latter is more effective for giving the composite stable hydrophobic properties. The composite material retains a highly hydrophobic state (θ ~ 142°), even after a long exposure to the atmosphere saturated with water vapor, it does not sink and does not change its shape after a prolonged stay in water (during the year) and is not subjected to destructive effects under a cyclic alternating temperature gradient freeze-thaw procedure. The mechanical strength of the composite was two times higher then that of the parent TZMK. In all cases the hydrophobic properties of aerogel-TZMK composite were higher then that of trimethylchlorosilane-treated TZMK. The created composite material exhibits properties that enable its use in Arctic, for which high humidity, low temperatures and the destructive effect of the freezing/thawing process are specific. It is logical to assume that these porous composites can form the basis for the creation of a new type of sorbents and catalysts - aerogel provides a high surface area and sorption properties, and TZMK gives the material the required mechanical strength.