See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/225174325

Influence of pore size distribution on water adsorption on silica gels

Article  in  Journal of Porous Materials · August 2010

DOI: 10.1007/s10934-009-9317-0

CITATIONS READS

17 1,265

3 authors:

J. Alcañiz-Monge María Pérez-Cadenas

University of Alicante National Distance Education University
65 PUBLICATIONS   3,144 CITATIONS    34 PUBLICATIONS   755 CITATIONS   

SEE PROFILE SEE PROFILE

Dolores Lozano-Castello
University of Alicante
121 PUBLICATIONS   6,950 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Special Issue of Catalysts (ISSN 2073-4344): Metal Supported Catalysts for Preferential Carbon Monoxide Oxidation (CO-PROX) View project

catalytic production of chemicals of energy interest from renewable materials View project

All content following this page was uploaded by María Pérez-Cadenas on 09 December 2014.

The user has requested enhancement of the downloaded file.

J Porous Mater (2010) 17:409–416
DOI 10.1007/s10934-009-9317-0

Influence of pore size distribution on water adsorption

on silica gels
J. Alcañiz-Monge Æ M. Pérez-Cadenas Æ
D. Lozano-Castelló

Published online: 5 August 2009

Ó Springer Science+Business Media, LLC 2009

Abstract The influence of pore size distribution on water characterization of these parameters is crucial for assessing
adsorption has been analysed for a series of partially their applicability for liquid or gas phase processes. Phys-
dehydroxylated silica gels with different pore structure, ical adsorption of gases is the most common technique for
assessed from the analysis of nitrogen and CO2 adsorption the characterization of porous solids [6–11]. Physical
isotherms. It has been observed that some silica gels show adsorption includes attractive dispersion forces and, at very
water adsorption mainly in the first range of relative short distances, repulsive forces, as well as contribution
pressures (relative pressure from 0 up to 0.6), while other from polarisation and electrostatic forces between perma-
silica gels also show water adsorption at higher relative nent electrical moments and the electrical field of the solid,
pressures. Comparison of the N2 and water adsorption if the adsorptive or the adsorbent have a polar nature.
isotherms points out that only silica gels containing mes- Depending on the nature of the molecule and the solid,
oporosity show water adsorption at relative pressures specific interactions may contribute to the adsorption
higher than 0.6. Moreover, a good correlation between the energy. Generally, non-polar molecules, like N2, CO2, Ar,
initial part of water adsorption isotherms (relative pressure He, CH4, benzene, nonane are used for this purpose [6–15].
from 0 up to 0.3) and CO2 adsorption isotherms has been Among the gases, N2 adsorption at 77 K is the most widely
found (the higher the CO2 adsorption capacity, the higher used and, usually, has a special status of recommended
the water adsorption capacity at low pressure). adsorptive [16]. On the other hand, polar molecules, like
NH3, H2O, alcohols are scarcely used for this purpose. This
Keywords Porous silica gels  Microporosity  is because the adsorption of polar molecules is more
Mesoporosity  Water adsorption complex than that of non-polar molecules [2, 17–19].
The mechanism of water adsorption on porous solids has
been the subject of numerous investigations during the last
1 Introduction several decades. However, at the present, several ambigu-
ities still remain unresolved, which make this subject very
The synthesis and structural characterisation of inorganic complex [9, 11, 20–22]. In previous studies, we analyzed
porous solids, like activated carbons (AC), silicas and the water adsorption in AC obtained from different pre-
zeolites, have received a growing and intensive research cursors. These materials covered a wide range of pore
during the last decades [1–5]. The parameters that are used volumes and had different contribution of mesoporosity
to define the pore structure of solids are apparent surface and microporosity [23–25]. It was shown that water
area, pore volume and pore size distribution and the adsorption on AC is influenced by the pore size distribu-
tion: water adsorbs in all the porosity, filling first the
microporosity, followed by the mesoporosity, and finally
J. Alcañiz-Monge (&)  M. Pérez-Cadenas  the macroporosity.
D. Lozano-Castelló
Water adsorption on amorphous silica has been also
Dpto. Quı́mica Inorgánica, Universidad de Alicante, Apdo. 99,
03080 Alicante, Spain investigated by numerous researchers due to the impor-
e-mail: [email protected] tance that water interaction with the surface of silica has on

123
410 J Porous Mater (2010) 17:409–416

many of the uses of this material, as a drying agent, catalyst micropore volume (pore size \ 2 nm), which contains the
support, chromatographic and filler [2–4, 9, 11, 26, 27]. volume of the narrow micropores and that of the supe-
However, those studies are mainly focused on the charac- rmicropores, was calculated applying the DR equation to
terization of the hydrophobicity degree of a given surface, the N2 adsorption data at relative pressures \0.14; and (iii)
or the structural stability toward water, while the effects of the volume of mesopores (2 nm \ pore size \ 50 nm) was
pore size distribution has not been analyzed in the litera- assessed by applying the BJH-method to the N2 adsorption
ture. From this starting point, the aim of the present work is data. The specific surface area was determined using the
to analyse the influence of pore size distribution on water BET (Brunauer–Emmett–Teller) equation.
adsorption on silica gels. Water adsorption studies were carried out at 298 K in an
automatic volumetric gas adsorption instrument (Belsorb
18). It should be noted that, in conventional gas adsorption
2 Experimental equipments is difficult to accurately measure P/P0 above
0.98, due to water condensation on the walls of the
A series of silica gels was synthesised from tetraethyl instrument. However, the instrument used in the present
orthosilicate (Si(OCH2CH3)4, TEOS) following the proce- study is design to avoid the condensation of water vapour
dure described in the literature [28]. A two-step reaction was on the walls of the apparatus. Moreover, this equipment
used in order to avoid formation of a precipitate. The prep- allows to measure for long times (40–80 h). Experimental
aration was done in a polypropylene beaker. For the prepa- data was corrected for adsorption on inner wall of appa-
ration of sample 3A, 39 mL ethanol, 32.6 mL water and ratus. Additionally, a blank experiment using the empty
50 mL Si(OCH2CH3)4 (250 mmol) were mixed, obtaining a bulb showed that water adsorption on the inner surface of
biphasic solution. 1.37 mL HCl (0.1 N) was added to this glass was negligible. Previous to water adsorption experi-
biphasic solution and after 25 min the solution became ments, the samples were degassed at 623 K under high
monophasic (pH 4). After 4 h of stirring, NH3 was added vacuum (10-6 Torr). Diffuse Reflectance Infrared Fourier
until pH 7 and a gel was formed within 5 min. The solid was Transform Spectroscopy (DRIFTS) characterisation of
dried under vacuum at 120 °C for 4 days. Changing the silica gels was carried out in a spectrometer Mattson
water/TEOS and ethanol/TEOS ratios three more samples Infinities (MI60). Powdered samples were located in a
(1A, 1B, 2B) with different porosity were obtained. Four controlled environment chamber available in this equip-
additional samples with different porosity were obtained by ment. Measurements were carried out under N2 flow
following the same procedure but adding NH3 until pH 10, (60 mL/min) and at temperature of 623 K.
instead of pH 7 (samples 2A, 3B, 4B, 5B). To make easier
the presentation and discussion of the results, the nomen-
clature of the samples in this paper is not related to the 3 Results and discussion
preparation conditions, but to their porosity. Thus, the
nomenclature of the samples includes a letter, which indi- 3.1 Porous texture characterisation of silica gels
cates the presence of mainly microporosity (denoted by
letter ‘‘A’’) or the existence of also mesoporosity (denoted Figure 1 contains the nitrogen adsorption isotherms corre-
by letter ‘‘B’’) and a number, which indicates the order of sponding to all the silica gels prepared in the present study.
samples from low to high total pore volume.
In order to minimize the influence of hydrophilic silanol
5B
groups on water adsorption on silica gels, prior to
400 4B
adsorption experiments, the major part of silanol groups 3B
existing on as prepared silica gel must be removed. Thus,
according with the literature [10], samples were degassed 300 2B
at 623 K under vacuum, down to 10-3 Torr. Porous texture
Vads (cc/g)

1B
3A
characterisation of these samples was done by gas 200 2A
adsorption (N2 and CO2 adsorption at 77 and 273 K,
respectively) (Quantachrome. Autosorb 6). The Dubinin– 1A
100
Radushkevich (DR) equation was used to calculate the
micropore volume. The pore volume distributions was
calculated as follows [11]: (i) the volume of narrow 0
0 0.2 0.4 0.6 0.8 1
micropores (\0.7 nm) was calculated by applying the
P/Po
Dubinin–Radushkevich (DR) equation to the CO2 adsorp-
tion data at relative pressures (P/P0) \0.015; (ii) the total Fig. 1 Nitrogen adsorption isotherms on a series of silica gels

123
J Porous Mater (2010) 17:409–416 411

45 3A Pore volume (cm3/g)

2A
40 1B 0.40
2B VN2 VCO2 Vmeso
35
3B 0.35
1A
30 4B
V ads (cc/g)

25 5B 0.30

20 0.25
15
0.20
10
0.15
5

0 0.10
0 0.005 0.01 0.015 0.02 0.025 0.03
P/Po 0.05

Fig. 2 CO2 adsorption isotherms on the series of silica gels 0.00

1A 2A 3A 1B 2B 3B 4B 5B

Fig. 3 Pore volume of narrow micropores (VCO2 ), total micropore

It is seen that samples 1A, 2A and 3A present Type I (VN2 ) and mesopore (Vmeso)
isotherms (according to the IUPAC classification), having a
considerable nitrogen uptake at low relative pressure, which
reflects the presence of microporosity [29]. On the other 1A with a low pore volume and narrow mean micropore
hand, samples 1B, 2B, 3B, 4B and 5B present, in addition to size (which is deduced from the fact that the two micropore
the uptake at low relative pressure, nitrogen uptake at rel- volumes are quite similar, VN2  VCO2 ) and (ii) samples
ative pressures P/P0 [ 0.2, approximately, which indicates 2A and 3A with a higher micropore volume and wider
the existence of mesoporosity. The different shape of the mean micropore size (VN2 [ VCO2 ). The next series of
nitrogen adsorption isotherms points out the different pore samples (B) contains both micropore and mesopore on
size distribution of this series of samples. Thus, in this series different ratio. It can be observed that, as mesopore volume
of samples there are essentially microporous samples increases, respect to VN2 , the narrow micropore volume
(without any mesoporosity), which is reflected by the pla- (VCO2 ) decreases, which indicates a clear relationship
teau of the isotherms (Series A), and samples containing between mesopore formation and widening of micropo-
both microporosity and mesoporosity (Series B), the last rosity. Thus, sample 5B shows the highest mesopore vol-
one reflected by the high slope of their isotherm and the ume, which is similar to the micropore volume (VN2 ), and
presence of a hysteresis loop. the lowest VCO2 of the series, pointing out that this sample
Figure 2 contains the CO2 adsorption isotherms at 273 K has the widest micropore size distribution.
for all the samples. It is known that CO2 adsorption at sub-
atmospheric pressure is a useful complementary technique 3.2 Water isotherms on silica gels: influence of pore
for the characterisation of the narrow microporosity, due to size distribution
the higher adsorption temperature (273 K) and the smaller
kinetic diameter of CO2 [11, 30, 31]. As it can be seen in Figure 4 presents water adsorption isotherms correspond-
Fig. 2, all silica gels present CO2 adsorption, indicating that ing to all silica gels. In this figure it can be observed that,
all these samples contain narrow microporosity. similarly to the N2 adsorption isotherms, silica gels of
For a better analysis of the pore volume distributions, Series A (essentially microporous) adsorb water in a unique
Fig. 3 contains the pore volume of different kinds of range of relative pressures, from low to medium relative
porosity calculated from the above described isotherms: pressures (0–0.6), while silica gels of Series B (containing
Narrow micropore volume from CO2 adsorption data simultaneously micro- and mesoporous) adsorb water in
(VCO2 ) (pore size \ 0.7 nm), total micropore volume from two ranges of relative pressures: the first one up to medium
N2 adsorption data (VN2 ) (pore size \ 2 nm), and meso- relative pressures, and the second one at higher relative
pore volume (Vmeso) (2 nm \ pore size \ 50 nm). Com- pressures.
parison of the values of VCO2 and VN2 gives us an idea From the comparative analysis of N2 and water
about the micropore size, while comparison of the values adsorption isotherms, it can be deduced that only silica gels
of VN2 and Vmeso gives an indication of the pore size dis- containing mesoporosity (series B) show water adsorption
tribution [11, 23, 24]. In Fig. 3 it is seen that in the series of on the range of high relative pressures. These results seem
samples prepared for this study, there are essentially to indicate that water adsorption at low relative pressures
microporous samples (without mesoporosity), with differ- occurs mainly on the microporosity, while the adsorption at
ent pore volume and micropore size distribution: (i) sample higher relative pressures seems to be due to the presence of

123
412 J Porous Mater (2010) 17:409–416

900 the specific contribution, which may be important to the

5B
800 overall adsorption energy according to the polar nature of
4B
the molecules and the adsorbents. In our case, water
700 3B
adsorption on silica gels, all terms must be considered, due
600 2B to both adsorbent and adsorptive show an important
V (cc/g) NTP

500 polarity. The non-specific contributions are weak in the

1B
3A case of water and silica pore wall, as a consequence of the
400
2A low polarizability of water molecule. Hence water
300 adsorption at low relative pressures must be very low
1A
200 according with this interaction.
The mechanism of water adsorption on porous solids is
100
very complex. It is known that surface of silica contains
0 silanol or siloxane groups [2]. The Si–OH shows a specific
0 0.2 0.4 0.6 0.8 1
interaction with water (hydrophilic), by means of the
P/Po
localised electron density of water molecule and the posi-
Fig. 4 Water adsorption isotherms at 298 K on silica gels tive charge of OH groups. On the other hand, Si–O–Si are
hydrophobic due to the non-polarity of these bonding
250
[34, 35]. Thus, it is reasonable to think that depending on
3A 2A 1B the amount of silanol groups in the surface of silica, the
200
2B amount of water adsorbed at very low relative pressures
1A would be different. In the case of partially dehydroxylated
V (cc/g) NTP

150
3B
silica, water adsorption is claimed to start by physical
4B adsorption: first water molecules bond to the remaining
100 5B hydroxyl groups on the surface of silica and, afterwards,
the growth of clusters of water molecules takes place
50
[10, 36, 37]; as the pressure increases, these clusters grow
and can coalesce to give a continuous adsorption layer on
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 the surface of the pores leading finally to the micropores
P/Po filling.
Considering all the results reported by other authors
Fig. 5 Water adsorption at low relative pressures
saying that a heat treatment up to 573 K on silica gels
removes a major part of silanol groups [10, 38], silica gels
mesoporosity. These results are in agreement with those of the present study must suffer an important dehydroxy-
obtained from water adsorption on AC [23, 24]. lation during the degassing process at 623 K, previous to
Next step in this research is to analyse the way water water adsorption experiments. This fact could explain the
adsorbs in each type of porosity. A deeper analysis of the low water uptake of these silica gels at low relative pres-
low relative pressure region (see Fig. 5), indicates that sures. These results agree with the study done by Baker and
water adsorption on silica gels is scarce at relative pressure Sing [39], where it was demonstrated that all dehydroxy-
lower than 0.1, compared with N2 adsorption. This is an lated silica presented a very small water adsorption at
indication of the weak water–silica interaction in this type P/P0 \ 0.3, due to the hydrophobicity of these samples.
of solids [32]. This is clear if we consider the overall Taking into account the above comments, the fact that
interaction potential energy (/T) for a polar adsorbate the initial adsorption is quite similar on all the studied
molecule on a heteropolar surface, which is given by the samples at very low relative pressures (0–0.025; see Fig. 5)
equation [8, 33, 34]: should be related to the interaction between water and
/T ¼ /D þ /P þ /F þ /FQ þ /R ; silanol groups. However, this cannot explain the different
water uptake at relative pressures higher than 0.025 that
where /D = energy due to dispersion forces; /P = energy follows the silica gels analysed. In order to analyse if there
due to polar nature of adsorbent and/or adsorptive; /F and is a relationship between the amount of hydroxo groups on
/FQ = energy due to permanent dipole and quadrupole silica gels and water adsorption capacity, the character-
moments of the adsorptive molecules, respectively; and isation of surface groups was investigated by infrared
/R = energy due to repulsion forces. The terms /D and /R spectroscopy (DRIFTS). Figure 6 contains the FTIR
are the non-specific contributions, which are always spectra corresponding to some selected silica gels. The
present in physisorption, whereas the other terms provide signal related to OH groups appears in the region between

123
J Porous Mater (2010) 17:409–416 413

highest N2 adsorption on the initial part of the isotherm and

2A the highest CO2 adsorption capacity (see Figs. 1 and 2,
respectively), which is indicating the presence of narrow
microporosity. The same correlation can be observed for the
Kulbelka-Munk

samples with the next highest water adsorption, samples 2A

3A and 1B. In fact, except from sample 1A, a good correlation
between the initial part of water isotherms (0–0.3) and their
CO2 isotherms and the initial part of N2 isotherms exists.
5B Sample 1A presents a similar water adsorption to sample 1B
on the range of relative pressures 0–0.2, while their CO2
1A isotherm and the initial part of N2 isotherm are comparable
to other samples (3A, 3B, 4B). According to these results,
3800 3400 3000 2600
sample 1A adsorbs more water at low relative pressures
Wave number (cm-1) than expected, which could be due to the presence of
micropores with very narrow size. In the very narrow pores
Fig. 6 DRIFT spectra of selected silica gels the adsorption potential is higher than in wider pores [40],
favouring the water adsorption at low relative pressures.
Thus, from all these results it can be said that, in a first step
3800 and 3000 cm-1. Each sample shows a sharp peak water fills the narrow microporosity. Samples 1A and 1B,
around 3750 cm-1 attributed to free OH groups, and a which have similar narrow micropore volume (VCO2 ; see
broad band between 3700 and 3000 cm-1, related to Table 1 and Fig. 2) show similar water adsorption in the
hydrogen-bonded OH groups. Comparing the intensity of range of P/P0 \ 0.3. The adsorption of water presented at
these bands and the water uptake at low relative pressure higher pressures must be related with the presence of wider
(Fig. 5), it can be observed that a correlation between the micropores (supermicropores; supermicropore volume =
amount of total OH groups and the amount of adsorbed VN2  VCO2 ). As microporosity gets wider, water adsorbs at
water at low relative pressure does not exist. Thus, it can be higher relative pressures (i.e. water adsorption in sample 3A
seen that the sample with the highest and lowest water ends at P/P0 around 0.6, while 1A ends at 0.5.
adsorption capacity (3A and 5B, respectively), present The existence of mesoporosity makes complex the
similar amounts of OH groups. Moreover, sample 1A, analysis of the transition from micropores to mesopores.
which contains a low amount of OH groups, has a quite Sample 3A, which does not present mesoporosity, as it can
high adsorption of water at low relative pressure. These be deduced from the small slope of the plateau in its N2
results indicate the small effect of surface chemistry in isotherm, adsorbs water mainly up to P/P0 around 0.6. On
water adsorption on partially dehydroxylated silica gels. the other hand, sample 2B, which has an important meso-
Considering the above comments, the isotherm shape porosity, shows a knee for water adsorption isotherm
must be a function of the micropore size distribution. Thus, around 0.7. Considering that the micropore volume for
the fact that sample 3A presents the highest water adsorp- both samples is quite similar (VN2 , Table 1), the higher
tion up to relative pressures of 0.4 (Fig. 5), agrees with the water adsorption from P/P0 around 0.6, for the sample 2B

Table 1 Pore volumes estimated from CO2, N2 and water adsorption data
Sample SBET VCO2 VN2 Vmeso Vtotal VLH2 O (0–0.95) VSH2 O (0–0.6) VLH2 O (0.6–0.95)
(m2 g-1) (cm3 g-1) (cm3 g-1) (cm3 g-1) (cm3 g-1) (cm3 g-1) (cm3 g-1) (cm3 g-1)

1A 354 0.16 0.20 0.00 0.20 0.19 0.20 0.02

2A 643 0.21 0.32 0.00 0.32 0.30 0.31 0.00
3A 640 0.23 0.34 0.00 0.34 0.32 0.34 0.02
1B 607 0.16 0.30 0.07 0.37 0.36 0.30 0.07
2B 689 0.15 0.37 0.10 0.47 0.45 0.37 0.10
3B 593 0.14 0.32 0.25 0.57 0.55 0.32 0.26
4B 646 0.13 0.32 0.31 0.63 0.61 0.33 0.30
5B 665 0.11 0.34 0.34 0.68 0.66 0.34 0.34
VCO2 = narrow micropore volume (pore \ 0.7 nm); VN2 = micropore volume (pore \ 2 nm); L = density of liquid water: 0.99 g cm-3;
S = density of solid water: 0.92 g cm-3

123
414 J Porous Mater (2010) 17:409–416

compared to 3A must be a consequence of water adsorption giving a value of 0.92 g cm-3, which is quite similar to the
on mesoporosity, which indicates that both processes ordinary solid phase of water [23–25]. This value has been
overlap. Thus, when a clear transition does not exist used to estimate the micropore volume from water
between micro and mesoporosity, as it is the case of these adsorption data up to P/P0 around 0.6, while the water
samples (Fig. 1), both the final filling of micropores and liquid density has been used to calculate the volume of
the start of mesopore filling could be taken place in the adsorbed water in the range of P/P0 around 0.6–0.95.
same range of relative pressures. Hence, they overlap given Table 1 contains these calculated values.
a single adsorption curve. As it can be observed in Table 1, the volumes obtained
The rest of mesoporous silica gels (3B, 4B, 5B) show from water adsorption are quite similar to the calculated
the knee of their water isotherm at higher relative pressures micropore and mesopore volume, which points out the
than the sample 2B. Sample 3B shows this knee around suitability of the approach considered in the present study.
P/P0 0.8, sample 5B around 0.9, and in the case of sample Water adsorbs initially on microporosity (up to P/P0 0.6),
4B the knee is less defined. This trend can be related to the as a solid-like structure. In these pores the density of
mesopore size distribution. From N2 isotherms, it can be adsorbed water can be considered to be similar to the ice
deduced that these samples present different mesopore size density. Recently, both Molecular Dynamics simulation
distribution. Sample 2B contains mesopores with narrow studies and experimental results of water adsorbed on silica
pore size, because N2 only adsorbs up to P/P0 0.6, while the pores gave a theoretical value of density of water equal to
others samples have wider mesoporosity because they 0.88 g cm-3 [22, 48]. On the other hand, the fact that the
present adsorption at higher P/P0. values of Vmeso and VLH2 O (0.6–0.95) included in Table 1
From these results it can be said that the shape of water are similar, points out that water adsorbs on mesopores as a
isotherm up to relative pressures around 0.6 is mainly liquid.
related to the micropore size distribution. On the other Water adsorption on the mesoporosity should take place
hand, the adsorption at higher P/P0 is due to the presence of mainly by the capillary condensation mechanism. Then, the
mesoporosity. use of Kelvin equation should be, in principle, correct to
describe this adsorption mechanism [10]. Using this
3.3 Water adsorption isotherms on silica gels: physical equation and taking into account the multilayer thickness
state of adsorbed phase of water adsorbed on silica, in previous works have esti-
mated that water adsorption on mesopores should start at
Another interesting result that can be extracted from water P/P0 around 0.45 [49, 50], which is not in agreement with
adsorption data is pore volume. Table 1 includes pore the experimental result obtained in the present work.
volume values considering that the adsorptive is in the However, it is important to note that none of those works
liquid state (liquid density for water 1.0 g cm-3). took into account the contact angle between water and the
According to the Gurvitsch rule, the pore volume calcu- surface of the silica. It was considered that water wets
lated from N2 adsorption data and water adsorption data completely the silica surface, despite all silica gels show,
should be similar [10]. However, it is usually observed that up to some extension, a hydrophobic character [51, 52].
pore volume estimated by water adsorption is lower than Thus, the Kelvin equation with correction for the contact
the nitrogen saturation volume [41, 42]. From this dis- angle h between liquid phase and pore wall should be used.
crepancy it has been suggested that the density of water in This equation is:
the adsorbed phase is lower than the normal liquid at the  
P0 2rVm cos h
corresponding temperature [10, 43]. Several studies have ln ¼
P RTrk
been carried out to understand the behaviour of water
confined into the pores. These include X-ray and neutron where r and Vm are, respectively, the surface tension and
diffraction studies, calorimetric (DTA, DSC) techniques the liquid molar volume of bulk water at temperature T, R
applied to the analysis of the freezing behaviour of the gas constant and rk is the core pore radius. For the case
adsorbed water, nuclear magnetic resonance (NMR) spec- of water adsorption at 298 K, the liquid molar volume is
troscopy, dielectric measurements and so on [44–46]. All Vm = 18 cm3 mol-1, and the surface tension is around
those studies showed that adsorbed water has different r = 72 mN/m. As it has been explained above, a critical
structures depending on the pore size where it is adsorbed. parameter is the contact angle, which has a great uncer-
On narrow pores (micropores) has a solid-like structure tainty, because liquid water wets the pore wall depending
even at room temperature, as a consequence of the strong on the groups existing on the surface of the silica: when
wall interaction. On the other hand, on mesopores, water is hydrophobic Si–O–Si groups exist, water cannot wet the
adsorbed as a liquid [47]. Experimental density of water surface; when hydrophilic Si–OH groups exist, water does
adsorbed on the microporosity of AC has been obtained, wet the surface. Thus, contact angle values are different

123
J Porous Mater (2010) 17:409–416 415

depending on the polarity of the surface. In the case of Acknowledgments The authors thank the Generalitat Valenciana
hydrophobic solids, as for example pure graphite, h value (project GV/2007/144 and PROMETEO/2009/047) and Ministerio de
Fomento (Project 70012/T-05) for the financial support.
equal to 86° was taken at 298 K [53]. However, the pres-
ence of polar compounds on the surface of carbon can
reduce the contact angle. Thus, even a little amount of References
these polar compounds causes a reduction of h, as it was
found for pyrolytic carbon (h equal to 72°) [53]. A high 1. R.C. Bansal, M. Goyal, Activated Carbon Adsorption (Taylor &
amount of polar compounds in the surface, as it is the case Francis, Boca Raton, FL, 2005)
of AC, produces a reduction of h (reported value of 54° on 2. R.K. Iler, The Chemistry of Silica (Wiley, New York, 1979)
3. H.E. Bergna, W.O. Roberts, Colloidal Silica: Fundamentals and
AC) [53]. In the case of silica gels, the contact angles Applications (Taylor & Francis, Boca Raton, FL, 2006)
varies from 95° (for hydrophobic porous silica) to 5° (for 4. J.A. Schwarz, C.I. Contescu, Surfaces of Nanoparticles and
hydrophilic silicas) [54, 55]. According to the low content Porous Materials (Marcel Dekker, New York, 1999)
of silanol groups on the silica gels of the present study due 5. R. Xu, W. Pang, J. Yu, Q. Huo, Chemistry of Zeolites and Related
Porous Materials: Synthesis and Structure (Wiley Interscience,
to the outgassing conditions, it is expected that these 2007)
samples present a similar behaviour to that of AC. Thus, 6. S. Lowell, J.E. Shields, Powder, Surface Area and Porosity, 3rd
using a value of 45° in Kelvin equation, water adsorption edn. (Chapman and Hall, Bristol, 1991)
on mesoporosity stars at P/P0 equal to 0.6, which agrees 7. H. Jankowska, A. Swiatkowski, J. Choma, in Active Carbon, ed.
by T.J. Kemp (Ellis Horwood, Chischester, 1991)
with the isotherm shape obtained in the present study. 8. S.J. Gregg, S.K.W. Sing, Adsorption, Surface Area and Porosity,
Water adsorption is different to N2 or benzene adsorption, 2nd edn. (Academic Press, London, 1982)
which wet completely the adsorbent (cos h = 1) and are 9. B. McEnaney et al., in Characterization of Porous Solids IV, ed.
not influenced by the presence of surface polarity. Hence, by B. McEnaney, T.J. Mays, J. Rouquérol, F. Rodrı́guez-Reinoso,
K.S.W. Sing, K.K. Unger (Royal Society of Chemistry, London,
due to this dependence of h on polar nature of samples, the 1997)
Kelvin equation can not be used for the analysis of pore 10. F. Rouquerol, J. Rouquerol, K. Sing, Adsorption by Powder and
size distribution from water adsorption isotherms. Porous Solids: Principles, Methodology and Applications (Aca-
demic Press Inc, Cornwall, 1999)
11. F. Rodrı́guez-Reinoso, A. Linares-Solano, in Chemistry and
Physics of Carbon, vol. 21, ed. by P.A. Thrower (Dekker, New
York, 1988), p. 2
4 Conclusions 12. K. Kaneko, N. Setoyama, T. Suzuki, in Characterization of
Porous Solids III, ed. by J. Rouquérol, F. Rodrı́guez-Reinoso,
From the comparative analysis of N2, CO2 and water K.S.W. Sing, K.K. Unger (Elsevier Science Publishers, B.V.,
Amsterdam, 1994), p. 593
adsorption isotherms, it has been concluded that water
13. N. Setoyama, M. Ruike, T. Kasu, T. Suzuki, K. Kaneko, Lang-
adsorption at low relative pressures occurs mainly in the muir 9, 2612 (1993)
microporosity, and only silica gels containing mesoporos- 14. K.A. Sosin, D.F. Quinn, J. Porous. Mater. 1, 111 (1995)
ity show water adsorption in the range of high relative 15. J. Rouquerol et al., in Characterization of Porous Solids III, ed.
by J. Rouquérol, F. Rodrı́guez-Reinoso, K.S.W. Sing, K.K. Unger
pressures. It has been seen that a good correlation between
(Elsevier Science Publishers, B.V., Amsterdam, 1994), p. 1
the initial part of water adsorption isotherms (P/P0 from 0 16. R.H. Bradley, B. Rand, J. Colloid Interface Sci. 169, 168 (1995)
to 0.3) and the CO2 adsorption isotherms exists. Thus, 17. K. Kaneko, K. Inouke, Carbon 8, 772 (1986)
water fills in the narrow microporosity, and as the micro- 18. P.J.M. Carrott, M.B. Kenny, R.A. Roberts, K.S.W. Sing, C.R.
Theocharis, in Characterization of Porous Solids II, ed. by F.
porosity gets wider, water needs higher relative pressure to
Rodrı́guez-Reinoso, et al. (Elsevier Science Publishers, B.V.,
adsorb. Amsterdam, 1991), p. 685
Pore volumes have been obtained from water adsorption 19. W. Rudzinski, R. Charmas, R. Borrowiecki, in Adsorption on
considering two different water densities. The density of New and Modified Inorganic Sorbents, vol. 99, Stud. Surf. Sci.
Catal., ed by A. Dabrowski, V.A. Tertykn (Elsevier Science,
water in solid phase (0.92 g cm-3) has been used to esti-
B.V., 1996), p. 357
mate the micropore volume from the amount of water 20. K.E. Collins, V.R. de Camargo, A.B. Dimiras, P.A. da Silva, C.H.
adsorbed up to relative pressure around 0.6. On the other Collins, J. Colloid Interface Sci. 291, 353 (2005)
hand, the density of liquid water has been used to calculate 21. J.S. Oh, W.G. Shim, J.W. Lee, J.H. Kim, H. Moon, G. Seo, J.
Chem. Eng. Data 48, 1458 (2003)
the volume of water adsorbed in the relative pressure range
22. V. Kochervitov, V. Alfredsson, J. Phys. Chem. C 111, 12906
around 0.6–0.95. The pore volumes obtained from water (2007)
adsorption data are quite similar to the corresponding 23. J. Alcañiz-Monge, A. Linares-Solano, B. Rand, J. Phys. Chem. B
micropore and mesopore volumes obtained by nitrogen 105, 7998 (2001)
24. J. Alcañiz-Monge, A. Linares-Solano, B. Rand, J. Phys. Chem. B
adsorption data, which seems to corroborate that water
106, 3209 (2002)
adsorbs in the microporosity as solid ice, while it adsorbs in 25. J. Alcañiz-Monge, D. Lozano-Castelló, Adsorpt.Sci. Technol. 21,
the mesoporosity as a liquid. 841 (2003)

123
416 J Porous Mater (2010) 17:409–416

26. D. Barby, in Characterisation of Powder Surfaces, ed. by G.D. 41. C.M. Bambrough, R.C.T. Slade, R.T. Williams, R.S.L. Burkett,
Parfitt, K.S.W. Sing (Academic Press, London, 1976) S.D. Sims, S. Mann, J. Colloid Interface Sci. 201, 220 (1998)
27. K.K. Unker, Porous Silica (Elsevier, Amsterdam, 1979) 42. R.H. Bradley, B. Rand, Carbon 29, 1165 (1991)
28. A.M. Buckley, M. Greenblatt, J. Chem. Edu. 71, 599 (1994) 43. D.C. Steytler, J.C. Dore, Mol. Phys. 56, 1001 (1985)
29. K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. 44. M.C. Bellissent-funel, R. Sridi-Dorbez, L.J. Bosio, J. Chem.
Pierotti, J. Rouquerol, T. Siemieniewska, Pure Appl. Chem. 57, Phys. 104, 10023 (1996)
603 (1985) 45. A. Watanabe, T. Iiyama, K. Kaneko, in International Symposium
30. D. Cazorla-Amorós, J. Alcañiz-Monge, A. Linares-Solano, of Carbon, Tokyo, 1998, p. 624
Langmuir 12, 2820 (1996) 46. K. Kaneko, J. Miyawaki, A. Watanabe, T. Suzuki, F. Mneunie,
31. A. Linares-Solano, C. Salinas-Martı́nez de Lecea, J. Alcañiz- Fundamentals of Adsorption (Elsevier, Amsterdam, 1998), p. 51
Monge, D. Cazorla-Amorós, Tanso 185, 316 (1998) 47. K. Morishige, K. Kawano, J. Chem. Phys. 110, 10 (1999)
32. A.V. Kiselev, Y.I. Yashin, Gas-Adsorption Chromatography 48. P. Gallo, Phys. Chem. Chem. Phys. 2, 1607 (2000)
(Plenum Press, New York, 1969) 49. H. Naono, M. Hakuman, J. Colloid Interface Sci. 145, 405 (1991)
33. K.S.W. Sing, in Characterization of Powder Surfaces, ed. by 50. A. Raoof, J.P. Guilbaud, H.V. Damme, P. Porion, P. Levitz, J.
G.D. Parfitt, K.S.W. Sing (Academic Press, London, 1975) Colloid Interface Sci. 206, 1 (1998)
34. G.J. Young, J. Colloid Sci. 13, 67 (1958) 51. R.S.H. Mikhail, F.A. Shebl, J. Colloid Interface Sci. 34, 65 (1970)
35. M.L. Hair, W. Hertl, J. Chem. Phys. 73, 4269 (1969) 52. M.N. Plooster, S.N. Gitlin, J. Phys. Chem. 75, 21 (1971)
36. L.J. Holysz, Mater. Sci. 33, 445 (1998) 53. A.W. Adamson, Physical Chemistry of Surfaces, 5th edn. (Wiley
37. A. Burneau, J. Lepag, G.J. Maurice, Non-Cryst. Solids 217, 11 Int. Pub, New York, 1990)
(1997) 54. M. Trau, B.S. Murray, K. Grant, F. Grieser, J. Colloid Interface
38. J. Mrowiec-Bialón, A.B. Jarzebski, L. Pajak, Langmuir 15, 6505 Sci. 148, 182 (1992)
(1999) 55. N.V. Churaev, M.J. Setzer, J.J. Adolphs, J. Colloid Interface Sci.
39. F.S. Baker, K.S.W. Sing, J. Colloid Interface Sci. 55, 605 (1976) 197, 327 (1998)
40. D.H. Young, A.O. Crowell, Physical Adsorption of Gases (But-
terworth, London, 1962)

123

View publication stats