Functionalized mesoporous materials for adsorption and release of different drug molecules: A comparative study

Article abstract of DOI:10.1016/j.jssc.2009.03.034

The adsorption capacity and release properties of mesoporous materials for drug molecules can be improved by functionalizing their surfaces with judiciously chosen organic groups. Functionalized ordered mesoporous materials containing various types of org

Full text of DOI:10.1016/j.jssc.2009.03.034

ARTICLE IN PRESS
Journal of Solid State Chemistry 182 (2009) 1649–1660
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
Functionalized mesoporous materials for adsorption and release of different
drug molecules: A comparative study
Ã
Gang Wang, Amy N. Otuonye, Elizabeth A. Blair, Kelley Denton, Zhimin Tao, Tewodros Asefa
Department of Chemistry, 111 College Place, Syracuse University, Syracuse, New York 13244, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The adsorption capacity and release properties of mesoporous materials for drug molecules can be
improved by functionalizing their surfaces with judiciously chosen organic groups. Functionalized
ordered mesoporous materials containing various types of organic groups via a co-condensation
Received 28 October 2008
Received in revised form
11 March 2009
synthetic method from 15% organosilane and by post-grafting organosilanes onto
a pre-made
Accepted 28 March 2009
Available online 9 April 2009
mesoporous silica were synthesized. Comparative studies of their adsorption and release properties
for various model drug molecules were then conducted. Functional groups including 3-aminopropyl,
3-mercaptopropyl, vinyl, and secondary amine groups were used to functionalize the mesoporous
materials while rhodamine 6G and ibuprofen were utilized to investigate the materials’ relative
adsorption and release properties. The self-assembly of the mesoporous materials was carried out in the
presence of cetyltrimethylammonium bromide (CTAB) surfactant, which produced MCM-41 type
materials with pore diameters of ꢀ2.7–3.3 nm and moderate to high surface areas up to ꢀ1000 m2/g.
The different functional groups introduced into the materials dictated their adsorption capacity and
release properties. While mercaptopropyl and vinyl functionalized samples showed high adsorption
capacity for rhodamine 6G, amine functionalized samples exhibited higher adsorption capacity for
ibuprofen. While the diffusional release of ibuprofen was fitted on the Fickian diffusion model, the
release of rhodamine 6G followed Super Case-II transport model.
Keywords:
Drug delivery
Drug release
Drug adsorption
Nanostructured drug delivery vehicles
Functionalized mesoporous materials
& 2009 Elsevier Inc. All rights reserved.
1. Introduction
controlled drug release properties can enhance the efficacy of
the drugs.
Since their discovery in the early 1990s [2,3], a class of
nanostructured materials called mesoporous silicates such as
MCM-41 have attracted the attention of many scientists as drug
delivery vehicles [4] because of their outstanding features such as
high surface area (typically 1000 m2/g), high porosity (typical pore
volumes of 0.5–1.5 cm3/g), well-ordered, tunable nanometer pores
(typically 2–15 nm pore diameter) [5–7] and ‘‘non-cytotoxic’’
properties [8,9]. In fact, different types of mesoporous silica
nanomaterials were already proved to be capable of carrying high
dosages of a variety of drugs in their mesopores [4,10–12].
Additional benefits of mesoporous silica materials for drug
delivery include the simplicity of tuning their pore sizes by
changing their templates in order to better accommodate drug
molecules of different sizes as demonstrated by extensive works
At present, the most common ways of delivering drugs to
humans are oral administration and injection. However, these
methods have lower efficiency for some therapies. Some ther-
apeutic agents are unstable or poorly soluble drugs; therefore,
new delivery systems are currently required. Functionalized
nanostructured materials are increasingly considered as great
candidates to make drug delivery vehicles and controlled drug
release systems. This is because they have suitable platforms that
can help minimize adverse reactions and unwanted side effects,
that many conventional drugs used today often pose [1]. With
some drug administration methods, the drugs have to often pass
through various physiological obstacles before they reach their
desired target, thus decreasing the amount of drug that gets to the
targeted site. The inability to deliver controlled therapeutic
concentration of drugs to the desired location can result in a
decrease in the efficacy of the drug [1]. Increasing the concentra-
tions of drugs to be delivered by using nanomaterial based
drug delivery vehicles with improved adsorption capacity and
´
by Vallet-Regı and co-workers [13] as well as other researchers
[13,14]. While smaller drug molecules and biomolecules can be
accommodated in mesoporous materials with smaller as well as
bigger pore sizes, larger drug molecules require materials with
bigger pore diameters [15]. Furthermore, mesoporous silica
materials contain residual silanol groups, that can further be
functionalized by different organic groups in order to modify their
surface properties [16,17]. This creates favorable surface–drug
interactions, which in turn result in improved adsorption capacity
Ã
Corresponding author. Fax: +1315 443 4070.
E-mail address: tasefa@syr.edu (T. Asefa).
0022-4596/$ - see front matter & 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2009.03.034

ARTICLE IN PRESS
1650
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
of the materials for drug molecules. Lin and co-workers have
shown that the organic functionalization of mesoporous materials
can also influence their biocompatibility [18]. In addition to
surface functional groups, the morphology and size of the
mesoporous materials also have an important influence on drug
release characteristics [19].
80 g/L in ethanol, 15 g/L in propanol and 100 g/L in diethylene
glycol. Rhodamine 6G has increased solubility at higher pH [26].
Ibuprofen, which is a relatively weak acid with pka value of 4.4,
has low solubility in water or at acidic pH. Ibuprofen has an
intrinsic solubility of ꢀ0.06 mg/mL in water. Ibuprofen is sparingly
soluble in hexane and freely soluble in ethanol, octanol and
dimethyl sulfoxide and chloroform with values of 410 g/L in
acetone, 410 g/L in ethanol, 33 g/L in octanol, and 3.3 g/L in
hexane. The solubility of ibuprofen increases sharply with pH, i.e.
the drug is largely insoluble at low pH, but is readily soluble at
alkaline pH. For example in water, its solubility is ꢀ0.5 ꢁ10ꢂ1g/L
at pHo2.00 but ꢀ1 ꢁ102 g/L at pH ¼ 7.5 [27]. Based on these
properties or because rhodamine 6G is quite soluble in water
while ibuprofen is rather soluble in solvents such as hexane and
ethanol, we have considered rhodamine 6G to be a hydrophilic
probe molecule while ibuprofen is considered hydrophobic in this
study. Therefore, they are expected to show different adsorption
and release properties in organic functionalized mesoporous
materials.
Our studies indicated that while the samples functionalized
with mercaptopropyl and vinyl groups resulted in high adsorption
capacity for rhodamine 6G, those functionalized with amine
groups showed higher adsorption capacity for ibuprofen. Simi-
larly, the drug release properties also varied from sample to
sample, depending on the type of functional groups they
contained. Furthermore, differences in adsorption capacity and
drug release properties between the materials synthesized via co-
condensation and those synthesized via post-grafting were also
observed. The results of our study may give further insights into
rational synthetic approaches to functionalized mesoporous
materials with improved adsorption capacity and release proper-
ties for a variety of hydrophobic and hydrophilic drugs.
Generally, surface functionalization of mesoporous silica
materials via covalent bonding of organic groups can be achieved
by two methods: i.e., post-grafting synthesis [20] and co-
condensation [20]. Although the post-grafting method results in
well-ordered functionalized mesostructured materials, it often
produces non-uniformly distributed organic groups because the
organic moieties can congregate more on the channel pore mouth
and on the exterior surfaces [21]. The Co-condensation synthetic
method of mesoporous materials involves a one-step procedure
and allows better control of the loading and distribution of the
organic groups [22] although it often produces materials with
less ordered mesoporous structures. In particular, low degree of
structural integrity and long-range periodicity as well as lower
surface area would be produced when the organosilane concen-
tration in the synthesis exceeds ꢀ25% [22].
Herein comparative investigations of the adsorption capacity
and drug release properties of mesoporous materials whose
surfaces are functionalized with judiciously chosen organic
groups via post-grafting or co-condensation of various organosi-
lanes were conducted. Furthermore, two different hydrophobic
and hydrophilic molecules were used as model drugs in the study.
The approach of organic functionalization of mesoporous materi-
als for drug delivery has been considered previously [23–25].
The most studied system is ibuprofen adsorption on organic-
functionalized matrices. When the MCM-41 or SBA-15 surface is
functionalized with amino groups [23c,d], there is an ionic
interaction between the carboxylate groups of ibuprofen and the
ammonium groups on the matrix surface. The hydrophobicity of
an ordered mesoporous silica can also be altered by modification
of the surface with alkyl chains [23e]. As a result, the hydrophobic
interaction with hydrophobic drugs can be improved. For instance,
erythromycin release from functionalized matrices was much
slower compared to unmodified material [23g]. Here a compara-
tive study between grafting and co-condensation as well as the
use of different functional groups and two different (model)
drug molecules have been performed. Functional groups including
3-aminopropyl, 3-mercaptopropyl, vinyl, and secondary amine
groups were used to functionalize the mesoporous materials
while rhodamine 6G (R6G) and ibuprofen were used as probe
molecules to investigate the materials’ adsorption and release
properties. The objectives of our study is to obtain the relative
effect of functional groups as well as type of synthetic method
to the functionalized materials on the adsorption and release
properties of the materials for different molecules. The self-
assembly of the mesoporous materials was carried out with
cetyltrimethylammonium bromide (CTAB) surfactant producing
MCM-41 type materials with pore diameters of ꢀ2.7–3.3 nm and
moderate to high surface areas up to ꢀ1000 m2/g. By changing the
organic groups, the properties of the mesoporous materials were
tuned from hydrophobic to hydrophilic and their adsorption and
release properties for different (model) drug molecules such as
rhodamine 6G and ibuprofen varied.
2. Experimental section
2.1. Materials and reagents
Tetraethoxysilane (TEOS), 3-aminopropyltriethoxysilane (APTS),
rhodamine 6G, cetyltrimethyammonium bromide (CTAB), ibupro-
fen sodium salt, 3-mercaptopropyltrimethoxysilane (MPTS), vinyl-
trimethoxysilane (VTS), NaCl, NaHCO3, KCl, K2HPO4 ꢃ 3H2O,
MgCl2 ꢃ 6H2O, CaCl2, Na2SO4, NH2C(CH2OH)3, and bis(triethoxysi-
lylpropyl)amine (BTSPA) were obtained from Sigma-Aldrich.
Hydrochloric acid (36.5%) and anhydrous toluene were purchased
from Fisher Scientific.
2.2. Synthesis of functionalized mesoporous materials
via co-condensation
A solution of 33.4 mL distilled water and 15 mL ammonium
hydroxide was prepared and 2.30 mmol CTAB was dissolved in it
by stirring. Then, a mixture of 17 mmol TEOS and 3.0 mmol of one
of the organosilanes (MPTS, VTS, APTS, or BTSPA) was added. The
solution was stirred at room temperature for 2 h and then stored
in oven at 80 1C for two days. The sample was cooled down and
filtered over Whatman-1 filter paper. The solid was washed
thoroughly with large amount of distilled water and dried under
ambient condition resulting in organic-functionalized mesostruc-
tured materials containing 3-mercaptopropyl, vinyl, 3-aminopro-
pyl, or N,N-diproplyamine groups, respectively. The surfactant
template was extracted by stirring 2 g of the functionalized
mesostructured material with a solution of 50 mL methanol and
10 mL HCl for 5 h at 50 1C. The solution was filtered over a
Whatman filter paper. The solid was washed three times with
Rhodamine 6G and ibuprofen were chosen as probe molecules
in our study because of their differences in hydrophilicity
(or hydrophobicity), which allows the investigation of interaction
of different molecules with functionalized mesoporous materials
[24]. Furthermore, they are easy to probe by UV–Vis absorption
spectroscopy. The solubility of rhodamine 6G and ibuprofen is
dependent on solvent and pH of solution. For instance, the
solubility of rhodamine 6G is 20 g/L in water; 40 g/L in butanol,

ARTICLE IN PRESS
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
1651
20 mL methanol and dried under vacuum for 30 min. This resulted
in organic-functionalized samples that were denoted as Co-MPTS,
Co-VTS, Co-APTS, and Co-BPSPA, which contained 3-mercapto-
propyl, vinyl, 3-aminopropyl, or N,N-diproplyamine groups,
respectively.
rhodamine 6G, as monitored by UV–Vis absorption spectroscopy.
The solution was filtered and the solid was washed with 20 mL of
water. This washing procedure was used to remove excess
rhodamine 6G molecules on the outer surface of the materials
[24g]. The amount of rhodamine 6G on the external area was
calculated to be 2.0–4.5% by subtracting the amount of rhodamine
6G removed in the washing steps from the total amount adsorbed
initially. The total loading of rhodamine 6G in the mesoporous
channels is calculated to be 30–55 mg for 100 mg material
(Table 2). Then, 100 mg the rhodamine 6G saturated sample was
mixed 50 mL SBF solution of pH 7.4 at 37 1C and stirred for 2 h.
The solution was centrifuged for 5 min and then the UV–Vis
absorption of 5 mL of the supernatant was tested by monitoring
the absorption maximum of rhodamine 6G at 534 nm. The SBF/
rhodamine 6G supernatant solution used for the measurement
was discarded and the same volume of a fresh SBF solution was
then added into the sample. After stirring and centrifugation of
the solution, by following the same procedure above, the super-
natant was analyzed again by UV–Vis spectroscopy. This experi-
ment was repeated until the release of rhodamine 6G by the
sample remained insignificant.
2.3. Synthesis of functionalized mesoporous materials via
post-grafting
First, mesoporous silica was prepared and then functionalized
with different organic groups. Typically, a solution of 960 mL
distilled water, 10.9 mmol CTAB and 40 mL NaOH (2.0 M) was
prepared at 80 1C. The solution was stirred at 80 1C for 30 min.
After adding 22.6 mL of TEOS, the solution was stirred for 2 h at
80 1C. It was then filtered over Whatman-1 filter paper and the
solid was washed thoroughly with copious amount of distilled
water, and dried under ambient condition. The surfactant
template was extracted by stirring 1 g of the functionalized
mesostructured material with
a solution containing 150 mL
ethanol and 0.6 mL HCl for 5 h at 50 1C to produce mesoporous
silica, MCM-41. The extracted MCM-41 was grafted with
3-mercaptopropyl, vinyl, 3-aminopropyl, and secondary amine
groups by stirring 600 mg MCM-41 with 4.4 mmol of MPTS, VTS,
APTS, or BTSPA, respectively, in 100 mL toluene for 5 h. The
samples were then washed with copious amount of ethanol and
let to dry. The MCM-41 to organosilane ratio of 600 mg/4.4 mmol
was chosen in order to ensure that excess organosilane is present
in the solution. Many of our and others’ previous studies [19] have
shown that this ratio is optimum for obtaining the maximum
possible grafted groups in the materials.
2.6. Adsorption of ibuprofen
With the same functionalized mesoporous materials, adsorp-
tion and release experiments with ibuprofen were performed.
First, ibuprofen was prepared by stirring 4.4 mmol of commer-
cially available ibuprofen sodium salt in 40 mL, 0.2 M HCl solution
overnight at room temperature. The ibuprofen was let to crystal-
lize. The solution was then filtered over Whatman-1 filter paper
and the solid ibuprofen was dried under vacuum for 30 min.
A solution of ibuprofen with a concentration of 2.2 mM in ethanol
was prepared. The UV–Vis absorption experiments were carried
out with this solution by following the procedure mentioned
above for rhodamine 6G but by monitoring the absorption
2.4. Adsorption of rhodamine 6G
Typically, a solution of 50
mM rhodamine 6G dye was prepared
in distilled water at room temperature. Then, 100 mg of the the
organic-functionalized mesoporous sample and 50 mL of the
rhodamine 6G solution were mixed and stirred. Every 30 min,
the solution was centrifuged for 5 min and then the supernatant
was analyzed with UV–Vis absorption spectroscopy, by monitor-
ing the major peak at 534 nm, which corresponds to the
absorption maximum of rhodamine 6G. The absorption measure-
ment was repeated, until the sample had stopped adsorbing more
dye or until the absorption spectra of the supernatant had barely
changed, indicating that saturation point was reached. Upon
nearly complete adsorption of rhodamine 6G from the super-
natant by the material, more rhodamine 6G solution was added
into the sample and the above procedure was repeated.
maximum of ibuprofen at 264 nm. To determine the
% of
adsorption of ibuprofen on the external surface of the materials,
50 mg of functionalized mesoporous sample was soaked in 50 mL
of 24.2 mM ibuprofen solution for 24 h. It was then filtered and
quickly washed once with 20 mL ethanol to remove excess
ibuprofen coating on the outer surface of the materials [24g].
The amount of ibuprofen on the external area was calculated by
subtracting the amount of ibuprofen removed in the washing
steps from the total amount adsorbed initially.
2.7. Ibuprofen release experiments
In another experiment, 50 mg sample was kept in different
concentrations of rhodamine 6G solutions for two days and after
centrifugation, the absorbance of the supernatant was measured.
This allowed us to plot % adsorption versus concentration, which
allowed us to determine the adsorption capacity of the samples.
Typically, 50 mg of the functionalized mesoporous sample that
was saturated with ibuprofen above was stirred in 10 mL of SBF
solution at pH 7.4 at 37 1C. After centrifugation of the sample for
5 min, 5 mL of the supernatant was tested with UV–Vis spectro-
scopy every 30–60 min to ensure that the mesoporous sample was
saturated with ibuprofen. The release experiments were generally
carried out by following the same procedure as the one carried out
for rhodamine 6G above, except monitoring the absorption
maximum of ibuprofen at 264 nm.
2.5. Release of rhodamine 6G
The experiments involving rhodamine 6G release by the
samples were performed in a simulated body fluid (SBF) solution
at 37 1C [16]. The SBF solution was prepared by mixing NaCl
(0.14 mol), NaHCO3 (4.20 mmol), KCl (3.00 mmol g), K2HPO4 ꢃ
3H2O (1 mmol), MgCl2 ꢃ 6H2O (1.50 mmol), 1 N HCl (40 mL),
CaCl2 (2.50 mmol), Na2SO4 (0.50 mmol), and NH2C(CH2OH)3
(49.90 mmol). A saturated sample with rhodamine 6G was first
prepared by mixing 100 mg of the functionalized mesoporous
2.8. Characterization
The UV–Vis absorption spectra were measured with a Lambda-
950 spectrophotometer (PerkinElmer). The thermogravimetric
(TGA) traces were collected by using a Q-500 Quantachrome
Analyzer (TA-Instruments) with N2 (99.999%) as a carrier gas
samples with 50 mL volume of 50
mM of rhodamine 6G solution
with
a heating ramp of 5 1C/min. The low angle powder
until the mesoporous sample had almost stopped adsorbing more
X-ray diffraction (XRD) patterns were obtained with a Scintag

ARTICLE IN PRESS
1652
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
Diffractometer. The BET gas adsorption–desorption measurements
were done with Micromeritics Tristar 3000 volumetric adsorption
analyzer, after degassing the samples at 160 1C for 12 h. The solid-
state 13C (75.5 MHz) and 29Si (59.6 MHz) NMR spectra were
acquired on a Bruker AVANCE 300 NMR spectrometer. For 13C
CP-MAS NMR experiments, 7.0 kHz spin rate, 5 s recycle delay,
VTS and MPTS showed high surface areas. The samples
synthesized from APTS and BTSPA by co-condensation showed
significantly lower surface areas. The pore size distributions of the
materials are shown in Fig. 3 and their average mesoporous
diameters are listed in Table 1. It is important to note here that the
absolute values of the pore diameters have to be treated with
caution as the BJH method, in general, underestimates the pore
diameter of mesoporous materials [31]. We also calculated
corrected values of pore diameters by using the equation
developed by Kruk and Jaroniec [31a,c]. The results reveal that
the materials synthesized by post-grafting have average mesopore
diameters between 2.7–3.2 nm while the samples synthesized by
co-condensation have average mesopore diameters of 3.0–3.3 nm
(Table 1). The average pore diameters of samples Co-VTS and Co-
MPTS were slightly lower than the corresponding samples
synthesized by post-grafting. This is consistent with previously
reported materials from co-condensation of hydrophobic
organosilanes, which showed smaller pore sizes and whose pore
sizes depend on the relative concentration of the organonosilane
in the solution [32]. Furthermore, upon immobilization of the
guest or drug molecules, the surface areas of the materials
decreased. For instance, the surface area of the Co-MPTS sample
decreased from 885 to 352 m2/g after getting saturated with
rhodamine 6G.
Typical 29Si CP-MAS NMR spectra for the functionalized
samples are shown in Supplementary materials (Fig. S1). The
spectra showed peaks at around –100 and –110 ppm which
correspond to hydroxyl containing silicon sites (Q3 or SiO1.5OH)
and cross-linked silicon (Q4 or SiO2), respectively. The spectra also
exhibited peaks between –50 and –80 ppm that were assigned to
the Si atoms covalently bonded to organic groups or T2
[(SiO)2Si(OH)(CH2–)] and T3 [(SiO)3Si(CH2)] sites [25]. The 13C
CP-MAS NMR spectra (Fig. S2) of APTS and MPTS functionalized
samples showed a signal at ꢀ10 ppm, which was assigned to
methylene, –CH2–, groups that are directly bonded to silicon
atom. The resonances between 20 and 30 ppm were attributed to
the other methylene and the terminal methyl groups of the propyl
chain [33]. Furthermore, Figs. S2,B, S2,C, and S2,D showed
additional minor peaks at ꢀ30 ppm that were assigned to carbon
atoms of residual CTAB surfactant in the pores of the samples.
These peaks were observed in the spectra for samples grafted with
MPTS, VTS and BTSPA while they were not visible on the spectrum
for the sample grafted with APTS (Fig. S1,A) although all these four
samples were prepared from the same batch of surfactant-
extracted MCM-41. This is because the density of organic groups
grafted in the former three samples was significantly lower than
that in the APTS-grafted sample. Thus, the residual surfactant
peaks appeared more visible with respect to the peaks corre-
sponding to the organic functional groups in the former than in
the latter.
1 ms contact time, p/2 pulse width of 5.6 ms, and 600 scans using
TPPM 1H decoupling were employed. For the 29Si CP-MAS NMR
experiments, 7.0 kHz spin rate, 10 s recycle delay, 10 ms contact
time, p/2 pulse width of 5.6 m
s, and 600 scans by using TPPM 1H
decoupling were employed.
3. Results and discussion
3.1. Synthesis and characterization of functionalized
mesoporous materials
The X-ray diffraction patterns of the functionalized mesopor-
ous materials and representative TEM images of samples
synthesized by post-grafting and co-condensation are shown in
Fig. 1. The X-ray diffraction patterns of all the samples synthesized
by post-grafting exhibited hexagonally ordered mesoporous
structure that is characteristic of MCM-41-type materials (Fig. 1, I).
The X-ray diffraction patterns (Fig. 1B) of the samples synthesized
by co-condensation from VTS and MPTS (Co-VTS and Co-MPTS)
also showed ordered mesostructures; however, those synthesized
from APTS (Co-APTS) and BTSPA (Co-BTSPA) revealed less ordered
mesostructures. The weakly ordered mesostructure in the latter
is probably a result of the fact that hydrophilic organoamine
groups in APTS and BTSPA cause some perturbation in the center
or the hydrophobic core of the surfactant micelles during co-
condensation synthesis of mesostructured materials from these
organosilanes [28]. From the (100) Bragg peak, the d-spacing
values of both Co-VTS and Co-MPTS samples as well as those
synthesized by post-grafting were calculated to be ꢀ4.3–4.5 nm.
Furthemore, except for a slight decrease in their (100) Bragg
reflection, the samples remained mesostructured with the same
unit cell size after adsorption and release of rhodamine 6G and
ibuprofen guest molecules (Fig. 1, II). The decrease in the intensity
of Bragg reflection is probably a result of a slight decrease in
electron contrast between the mesopores and the mesopore
channel walls due to the confinement of the R6G and ibuprofen
guest molecules inside the mesopores [17]. Although the unit
cells remained the same, the pore diameters and surface areas
of the materials decreased after adsorption of the drug or guest
molecules.
The morphology and mesopores of the materials are shown in
Fig. 1, III. Although the morphology and size of the mesoporous
materials are known to have an important influence on drug
release characteristics [29,30], the materials we used for our
comparative studies are made from the same batch of parent
material, and consequently had similar morphologies. This is,
therefore, expected to cause little difference on the adsorption and
release properties of the materials in our studies.
The mesoporous materials were further characterized
by thermogravimetric analysis under nitrogen (Table S1 and
Fig. S3) and elemental analysis (Table 2). These results further
corroborated the results obtained from solid-state NMR.
Thermogravimetric analysis of organic functionalized mesoporous
materials under nitrogen or air were previously used by many
groups for characterization of organic functionalized mesoporous
materials [34]. The TGA traces of the functionalized samples
(Fig. S3) showed a weight loss below 100 1C due to physisorbed
water [35] and, most importantly, a weight reduction between
ꢀ200 and 600 1C due to the loss of organic functional groups.
Furthermore, a slight weight reduction after 600 1C due to loss
of water molecules from condensation of residual silanols was
observed. Elemental analysis revealed that the mmol organic/g of
sample synthesized by post-grafting of APTS, BTSPA, MPTS, and
VTS were 2.17, 0.94, 0.97, and 1.53 mmol/g, respectively (Table 2).
The structures of the functionalized samples were also
characterized by nitrogen gas adsorption. The gas adsorption
(Fig. 2) showed
a type IV isotherm with sharp capillary
condensation steps for all the samples synthesized by post-
grafting as well as for samples Co-MPTS and Co-VTS that were
synthesized by co-condensation. This is indicative of the presence
of mesoporous structure in these materials. This result is also
consistent with their XRD patterns shown in Fig. 1. The specific
surface area and pore sizes of the materials are reported in Table 1.
It is worth noting that all the samples synthesized by post-
grafting and the samples synthesized by co-condensation from

ARTICLE IN PRESS
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
1653
(100)
15000
12000
9000
6000
3000
0
Grafting-APTS
Grafting-BTSPA
Grafting-MPTS
Grafting-VTS
(110)
(200)
1
2
3
4
5
6
7
2θ (Degrees)
3500
3000
2500
2000
1500
1000
500
(100)
2500
2000
1500
1000
500
(100)
C
A
B
D
4
0
0
0
2
4
6
8
0
2
6
8
2θ (degrees)
2θ (degrees)
Fig. 1. (I) Powder X-ray diffraction (XRD) patterns of functionalized mesoporous samples synthesized by post-grafting of organosilanes MPTS, VTS, APTS and BTSPA onto
mesoporous silica. (II) Powder X-ray diffraction (XRD) patterns of functionalized mesoporous samples synthesized by co-condensation of organosilanes (A) MPTS and (C)
VTS with the corresponding XRD patterns of the samples after R6G adsorption shown in (B) and (D), respectively. (III) Low resolution and high resolution TEM Images of
mesoporous samples synthesized by (A,B) post-grafting and (C,D) co-condensation.
This indicated that the surface density of functional groups
that were grafted decreased in the order of APTS 4VTS
4MPTSffiBTSPA. On the other hand, the mmol organic/g sample
in the samples synthesized by co-condensation, Co-MPTS and Co-
VTS, were 2.18 and 2.57 mmol/g, respectively (Table 2). This
indicates that the mmol of functional groups in the samples
is higher for sample prepared from VTS than that from MPTS
(or Co-VTS 4Co-MPTS). We can also conclude from Table 2 that

ARTICLE IN PRESS
1654
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
6.0
4.5
3.0
16
12
8
Grafting-APTS
Grafting-BTSPA
Grafting-MPTS
Grafting-VTS
Extracted-MCM41
Grafting-APTS
Grafting-BTSPA
Grafting-MPTS
Grafting-VTS
1.5
0.0
4
0.0
0.3
0.6
P/Po
0.9
20
30
40
50
60
Pore Size (Å)
24
18
12
6
Co-APTS
Co-BTSPA
Co-MPTS
Co-VTS
Co-APTS
Co-BTSPA
Co-MPTS
Co-VTS
2
1
0
0
20
40
60
80
100
0
Pore Size (Å)
0.0
0.3
0.6
0.9
Fig. 3. Pore size distribution of functionalized samples synthesized by (A) post-
grafting APTS, BTSPA, MPTS, and VTS, onto MCM-41 and (B) co-condensation of
APTS, BTSPA, MPTS, and VTS with TEOS.
P/Po
Fig. 2. Nitrogen adsorption isotherms of samples synthesized by (A) post-grafting
and (B) co-condensation of different organosilanes.
Table 1
was calculated. Although there was no significant difference in
surface area and pore size (Table 1) between the four samples
synthesized by post-grafting, the samples functionalized with
vinyl and mercaptopropyl groups adsorbed significantly more
rhodamine 6G than the samples functionalized with primary and
secondary amines (Table 2 and Fig. 4). This trend was also true for
samples synthesized by co-condensation. Upon comparing
samples synthesized under similar conditions, the samples
containing mercaptopropyl groups were found to adsorb the
most rhodamine 6G than the sample containing vinyl groups,
which in turn adsorbed more than the amine-functionalized
sample (Table 2). Generally, the adsorption capacity values for
rhodamine 6G appeared to be the highest in the sample
containing mercaptopropyl groups than vinyl groups and in
samples synthesized by co-condensation than by post-grafting
(Table 2). The adsorption capacity of the amine functionalized
samples for rhodamine 6G was significantly lower. Careful
analysis of the adsorption capacities by normalizing them with
respect to the density of functional groups in the materials as well
as their surface areas were conducted (Table 2). It indicated that
the samples containing mercaptopropyl groups with grafting
and co-condensation synthesis gave values of 1.5 ꢁ10ꢂ4 and
3.3 ꢁ10ꢂ4 mmol/m2, respectively, while those containing vinyl
groups with grafting and co-condensation gave 4.9 ꢁ10ꢂ5 and
6.8 ꢁ 10ꢂ5 mmol/m2, respectively. The samples synthesized
from APTS and BTSPA by grafting gave values of 1.1 ꢁ10ꢂ7
and 2.3 ꢁ10ꢂ7 mmol/m2, respectively. This indicates that the
mercaptopropyl groups results in an order of magnitude higher
adsorption capacity than those containing vinyl groups, which in
Structural data of organic functionalized mesoporous materials.
Samples
Surface
area
Average BJH
mesopore
Average mesopore
diameter by KJS
a
b
(m2/g)
diameter (A)
method (A)
˚
˚
Co-APTS
189
238
885
1183
887
852
952
811
2.4
2.3
2.2
2.4
2.0
2.0
2.1
2.2
3.1
3.2
2.7
3.0
3.0
3.0
3.2
3.3
Co-BTSPA
Co-MPTS
Co-VTS
Grafting-APTS
Grafting-BTSPA
Grafting-MPTS
Grafting-VTS
a
The pore diameters were obtained by the BJH method, which is known to
underestimate the pore diameter of mesoporous materials (see Ref. [31]). It is,
therefore, worth noting that the absolute values of these pore diameters have to be
treated with caution.
b
The pore diameters were calculated by an equation developed by Kruk and
Jaroniec [31a,c].
the sample synthesized from MPTS and VTS by co-condensation
had more functional groups compared to the corresponding
sample synthesized by post-grafting.
3.2. Adsorption and release of rhodamine 6G
By using the Beer–Lambert law, the amount of molecules
adsorbed in the mesoporous samples or released by the samples

ARTICLE IN PRESS
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
1655
Table 2
Density of grafted organic groups (mmol/g) of functionalized mesoporous materials determined from elemental analysis and adsorption capacity of the samples for
rhodamine 6G.
Samples
Grafting APTS Grafting BTSPA Grafting MPTS Grafting VTSa Co-MPTS
Co-VTSa
mmol R6G adsorbed/g sample
1 ꢁ10ꢂ4
2.17
2 ꢁ10ꢂ4
0.94
0.14
0.04
0.29
2.18
0.08
2.57
mmol Organic groups/g sampleb
0.97
1.53
Average number of adsorbed molecules per one surface ligand
Normalized with surface area (mmol R6G adsorbed/surface area, mmol/m2) 1.1 ꢁ10ꢂ7
4.6 ꢁ10ꢂ5
2.1 ꢁ10ꢂ4
1.4 ꢁ10ꢂ1
2.6 ꢁ10ꢂ2
1.3 ꢁ10ꢂ2 3.1 ꢁ10ꢂ4
2.3 ꢁ10ꢂ7
1.5 ꢁ10ꢂ4
4.9 ꢁ10ꢂ5
3.3 ꢁ10ꢂ4 6.8 ꢁ 10ꢂ5
a
mmol/g the samples functionalized with VTS could be slightly off as they were calculated based on the wt% of C and as the samples have some additional carbon from
the residual surfactant.
b
Determined from elemental analysis.
turn showed two orders of magnitude higher than the amine
functionalized samples. The rate of adsorption seems to be very
similar in all the samples in the initial period (for ꢀ11 h). The
samples functionalized with VTS by post-grafting as well as by co-
condensation almost got saturated within the first 11 h. However,
the samples functionalized with MPTS continued to adsorb more
rhodamine 6G for more than 11 h.
The adsorption profiles plotted by measuring adsorption
versus concentration are shown in Fig. 4C. All the lines were fit
into one type of function, Y ¼ A(1ꢂeꢂbX), where A and b are
constants (Table 3). The term A represents the maximum drug
adsorption by the materials while the value of (bA) represents the
initial adsorption rate of the drug by the nanoparticles and the
term (bAeꢂbX) indicates that the adsorption rate decreases over
time. Analysis of these terms A and b indicates that MPTS-
functionalized samples have higher adsorption capacity than
their corresponding VTS-functionalized samples in both cases,
i.e. samples synthesized by grafting and co-condensation.
Furthermore, the data shows that the initial adsorption rates of
the samples decrease in the order of Co-MPTS 4Grafting-MPTS
4Co-VTS ¼ Grafting-VTS.
MPTS and VTS appeared deep pink colored. This indicates that the
samples functionalized with APTS and BTSPA either by post-
grafting or co-condensation adsorbed little rhodamine 6G while
the samples functionalized with MPTS and VTS by both methods
adsorbed greater amounts of R6G.
Comparison of the density of functional groups in the materials
with respect to their adsorption properties to R6G reveals an
interesting trend. Although the density of grafted groups
in APTS functionalized sample is higher than that in the MPTS
functionalized sample, the former has a lower degree of adsorp-
tion of R6G. The adsorption capacity of the samples was normal-
ized with the density of organic functional groups in the materials
and their surface areas. It reveals that the differences in
adsorption properties of the samples to rhodamine 6G could be
partially attributed to the differences in the type of the functional
groups and possible degree of interaction between the functional
groups and R6G molecules. Recently Suh et al. [23a] have made
extensive studies in this area, which indicated that rhodamine
based dyes with positive charges such as rhodamine 6G and
rhodamine 123 show faster uptake than those with negative
charged such as rhodamine 101. Their results suggested that the
molecular charge of the drug has an important effect for the
uptake rate of the drugs in the materials. This molecular-charge
effect is the result of negative charges that are developed in the
pore surfaces by the ionization of silanol groups in aqueous
solution and their interactions with the charges of the incoming
dye molecules. Among the positively charged dyes, both the
molecular weight and the magnitude of the dipole moment seem
to contribute to the diffusion kinetics of the dyes. Their study and
Sekine and Nakatani [23f] have also reported uptake half-lives of
10–30 min at room temperature depending on the solution pH,
the ionic strength, the type of material (powder versus thin film)
as well as the channel length and pore structures. We observed
that the uptake half-times are slightly longer than what was
observed by others, with the MPTS-functionalized samples
reaching to uptake half-lives more slowly than the VTS-functio-
nalized samples. However, the former gave higher overall uptake
of rhodamine 6G than the corresponding VTS-functionalized
sample. Generally, the MPTS- and VTS-functionalized samples
gave higher adsorption capacity than MCM-41 and amine-
functionalized samples. The functional groups such as mercapto-
propyl and vinyl, which are hydrophobic, favorably interact with
the hydrophobic parts of rhodamine 6G, producing higher
adsorption capacity to R6G compared to MCM-41 and the
materials functionalized with organoamines. It appears that the
hydrophilic organoamines, –NH2 and –NH– groups, in the APTS
and BTSPA functionalized materials, respectively, produce un-
favorable interaction with the aromatic and alkyl chains of
rhodamine 6G molecule and thus, result in lower adsorption
capacity to R6G. In fact, the amine functional groups would likely
to be protonated by abstracting a proton from the silanol groups
[36]. This then leads to samples that have a positively charged
The adsorption results shown in Fig. 4 were further compli-
mented by comparing the weight losses between ꢀ150 and 700 1C
on thermogravimetric analysis (TGA) of the samples before and
after adsorption of R6G. The weight loss between 150 and 700 1C
can correspond to the decomposition of R6G or drug molecules.
Fig. 5 shows the TGA traces of the samples after incubation in
50 mM R6G solution (or the samples shown in Fig. 4D). The TGA
traces of sample Co-MPTS exhibited a 15.2% weight loss before
adsorption of R6G adsorption and a 23.4% weight loss after
adsorption of R6G (Fig. 5I). This corresponds to adsorbed R6G
weight of 8.2% and consistent with the adsorption capacity
obtained in the adsorption profile in Fig. 4D. The TGA traces of
sample Co-VTS exhibited a 5.1% weight loss before rhodamine 6G
adsorption and a 10.0% weight loss after R6G adsorption (Fig. 5II).
This corresponds to R6G weight of 4.9%, which is also consistent
with the adsorption capacity the sample obtained in Fig. 4D. The
increase in the values of weight loss after adsorption of R6G
clearly suggests that R6G was adsorbed in the samples and the
difference between the two values corresponds to the amount of
R6G adsorbed. These data indicate again that Co-MPTS adsorbed
more R6G than Co-VTS and is consistent with the data obtained
from the UV–Vis spectra analysis.
The difference in the degree of adsorption of R6G by the
samples can aso be clearly seen by simply looking at the color
of the mesoporous materials after treatment with R6G solution
(Figs. 5D and E). After mixing the same mass mesoporous
materials with the same volume and concentration of R6G
solution for several hours and then separating the solid from
the supernatant, samples with colors ranging from white to dark
pink were obtained. While the samples functionalized with APTS
and BTSPA almost still looked white, those functionalized with

ARTICLE IN PRESS
1656
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
3
2.5
2
1.5
1
0.5
0
Rhodamine 6G
MPTS
VTS
APTS
AAPTS
400
450
500
550
600
400
450
500
Wavelenght (nm)
550
600
wavelength (nm)
0.15
0.12
0.09
0.06
0.03
0.00
60
50
40
30
20
10
0
Grafting-MPTS
Grafting-VTS
Co-MPTS
Co-VTS
Grafting-MPTS
Grafting-VTS
Co-MPTS
Co-VTS
0
4000
8000
12000
16000
0
10
20
30
t (h)
40
50
60
μ
Ceq ( mol/L)
MPTS
VTS
APTS
BTSPA
Fig. 4. Adsorption properties of rhodamine 6G in various functionalized mesoporous materials. (A) UV–Vis absorption spectra of R6G in the supernatant 20 min after
adsorption of a solution of R6G with different functionalized mesoporous materials synthesized via co-condensation. (B) UV–Vis absorption spectra of R6G in the
supernatant 20 min after adsorption of R6G with different functionalized materials that were synthesized via post-grafting. (C) Adsorption versus concentration of R6G for
samples synthesized by post-grafting and co-condensation. (D) Percent of R6G adsorbed versus time profile of different samples synthesized by post-grafting and co-
condensation using aqueous 50 mM rhodamine 6G solution. (E) Digital images showing colors of the functionalized mesoporous samples synthesized by co-condensation
after adsorption of R6G. A bright color was observed for MPTS functionalized sample indicating that it has adsorbed the most R6G. (F) Digital images showing colors of the
functionalized mesoporous samples synthesized by post-grafting after adsorption of R6G.
surface which is not conducive for adsorption of positively
charged R6G molecules.
The trend in the degree of R6G release (Fig. 6) by the
functionalized samples was found to be the reverse of the one
observed for adsorption. The studies of R6G release were
conducted in a simulated body fluid solution of pH of 7.40 at
37 1C [16]. SBF solution was chosen as it has similar composition
as what is in our body and as it helps understanding how drug

ARTICLE IN PRESS
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
1657
Grafting MPTS
100
Table 3
Values of terms A and b in the graphs for adsorption versus concentration of
rhodamine 6G that are fit into a function: Y ¼ A(1ꢂeꢂbX), where A and b are
constants for each specific adsorption.
Grafting VTS
Co-MPTS
Co-VTS
80
60
40
20
0
Particles
A (%)a
b (
m
molꢂ1L)b
r2
Co-MPTS
43
40
55
30
5 ꢁ10ꢂ4
3 ꢁ10ꢂ4
2 ꢁ10ꢂ4
4 ꢁ10ꢂ4
0.98
0.99
0.98
0.99
Co-VTS
Grafting-MPTS
Grafting-VTS
a
The term A represents the maximum drug adsorption by the materials.
b
The term (bA) indicates the initial adsorption rate of the drug by the
nanoparticles while the value of (bAeꢂbX
)
indicates that the adsorption rate
decreases over time.
0
20
40
60
80
t (h)
A-Unextracted Co-MPTS
B-Extracted Co-MPTS
C-Extracted Co-MPTS-R6G
Fig. 6. Comparative release profiles percent of R6G released for different samples
synthesized by post-grafting and co-condensation of MPTS and VTS. The graphs
were fitted as function Y ¼ atn , where Y is % Release and t is time. This indicated
that for samples grafted with MPTS, a ¼ 0.0011, n ¼ 2.6, r2 ¼ 0.991; grafted with
100
90
VTS, a ¼ 0.16, n ¼ 1.7, r2 ¼ 0.991; sample Co-MPTS, a ¼ 1.14 ꢁ10ꢂ5
r2 ¼ 0.990; and sample Co-VTS, a ¼ 9.1 ꢁ10ꢂ4, n ¼ 2.9, r2 ¼ 0.987.
,
n ¼ 3.2,
released R6G faster than the corresponding samples functionalized
with MPTS.
80
B
C
The diffusional R6G release data was fitted onto a non-Fickian
or
a Super Case-II transport model [37,38] and followed
70
diffusional release process that follow Eq. (1) below with n 4 1:
Y ¼ atn
(1)
60
A
where Y is % Release and t is time. Further analysis of the data and
the graph in Fig. 6 gave values of a ¼ 0.0011, n ¼ 2.6, r2 ¼ 0.991 for
sample grafted with MPTS; a ¼ 0.16, n ¼ 1.7, r2 ¼ 0.991 for sample
grafted with VTS; a ¼ 1.14 ꢁ10ꢂ5, n ¼ 3.2, r2 ¼ 0.990 for sample
0
100 200 300 400 500 600 700
T ( C)
°
Co-MPTS synthesized by co-condensation; and a ¼ 9.1 ꢁ10ꢂ4
,
n ¼ 2.9, r2 ¼ 0.987 for sample Co-VTS synthesized by co-con-
A-Unextracted Co-VTS
B-Extracted Co-VTS
C-Extracted Co-VTS-R6G
densation.
100
90
3.3. Adsorption and release studies for ibuprofen
B
The adsorption experiments with ibuprofen (Fig. 7) exhibited a
totally different trend compared to the results obtained for
adsorption of R6G for the same series of samples. The UV–Vis
absorption data showed that the samples functionalized with
APTS and BTSPA by post-grafting had much more adsorption
capacity for ibuprofen than for R6G. Furthermore, the sample
grafted with APTS showed higher adsorption capacity of ibuprofen
than the corresponding sample grafted with BTSPA (Fig. 7A). On
the other hand, the samples synthesized from MPTS and VTS,
either by co-condensation or grafting, and the parent MCM-41
showed almost no adsorption of ibuprofen (Figures not shown).
The UV–Vis absorption results were corroborated by TGA data
(Table 4 and Fig. 7B). These results are interesting considering the
fact that elemental analysis (Table 2) shows the BTSPA-grafted
sample has almost similar mmol of grafted groups as the MPTS-
grafted sample but much less than the VTS-grafted sample and
the APTS-grafted sample has less mmol of functional groups
than the VTS-grafted sample. This further indicates that the type
of functional groups (or the degree of their interaction with
ibuprofen) and not their density in the materials is mainly
responsible for the differences in adsorption capacity of materials.
Since ibuprofen contains a carboxylic acid group, these carboxylic
acid groups could interact favorably with the –NH2 and –NH–
groups in the amine-functionalized samples via hydrogen bond-
ing, producing higher adsorption capacity of these samples for
ibuprofen. On the other hand, the hydrophobic mercaptopropyl and
C
80
70
A
0
100 200 300 400 500 600 700
T (°C)
Fig. 5. Thermogravimetric (TGA) traces of before surfactant extraction (A), after
surfactant extraction (B), and after R6G adsorption (C) for (I) samples synthesized
by co-condensation of MPTS and (II) samples synthesized by co-condensation of
VTS.
release properties of the materials would be in our body.
The samples functionalized with VTS by post-grafting method
released R6G molecules faster than the corresponding sample
synthesized by co-condensation. On the other hand, the sample
synthesized by post-grafting with MPTS released R6G mole-
cules faster than the corresponding sample synthesized by co-
condensation, therefore, the samples functionalized with VTS

ARTICLE IN PRESS
1658
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
0.12
0.09
0.06
0.03
0.00
100
80
Grafting APTS
Grafting BTSPA
Amount Adsorbed in 50 mg Sample, Used for this Study
60
40
20
0
APTS: 30.2
BTSPA: 33.2
Grafting-APTS
Grafting-BTSPA
0
4
8
12
16
Time (h)
Fig. 8. (A) Release of ibuprofen in SBF solution by different samples functionalized
with APTS and BTSPA. The graph for sample grafting APTS was fitted as Y ¼ kt1/2
(to2.6), Y ¼ 100 (t4 2.6), r2 ¼ 0.996 and the graph for sample grafting BTSPA was
fitted as Y ¼ kt1/2 (to2.6), Y ¼ ꢀ100 (t42.6), r2 ¼ 0.996.
0
5
10
t (h)
15
A – Grafting-BTSPA
B – Grafting-APTS
Table 5
100
95
90
85
80
75
70
Date of ibuprofen release shown in Fig. 8 for APTS and BTSPA grafted samples after
the graphs are Y ¼ kt (if tot*) or Y ¼ A (if tXt*), where Y represents percentage of
ibuprofen released, and t is time.
C – Grafting-BTSPA-Ibuprofen
D – Grafting-APTS-Ibuprofen
Samples
k (hꢂ1
)
t* (h)
A (%)
r2
A
B
Grafting-APTS
Grafting-BTSPA
34.2
32.9
2.9
3.3
100.3
108.6
0.95
0.94
C
D
t* is a time constant when the release reaches equilibrium (A, i.e., the maximum
release), whereas k stands for the rate at which ibuprofen gets released by the
materials.
25
225
425
T (°C)
625
825
stands for the rate at which ibuprofen gets released by the
particles. After peak fitting the data, we obtained n ¼ 1 for
tot*, where t* is a time constant when the release reaches
equilibrium (A, i.e., the maximum release), and has values of 2.9
and 3.3 h for APTS and BTSPA grafted samples, respectively (Eq. 2)
(Table 5):
Fig. 7. Adsorption properties of ibuprofen by various amine functionalized
mesoporous materials. (A) Percent of ibuprofen adsorbed versus time profile of
samples synthesized by post-grafting. (B) Thermogravimetric analysis of mesopor-
ous materials before and after adsorption of ibuprofen.
Y ¼ kt
(2)
and n ¼ 0 for tXt* (Eq. (3))
Y ¼ 100
Table 4
Percent weight loss between 200 and 600 1C for samples grafted with APTS and
BTSPA, before and after adsorption of ibuprofen.
(3)
As can be seen above, the diffusional release of the two molecules,
R6G and ibuprofen, from the functionalized materials follows
different mechanisms. This may have to do with the fact that the
interaction of R6G with the organic functional groups is
dominated by hydrophobic interaction while the interaction
between ibuprofen and the organoamine and residual silanol
groups of the functionalized mesoporous materials is dominated
by hydrogen bonding. Suh et al. [23b] reported that the control of
the inclusion and release characteristics of drug molecules in
mesoporous materials by manipulating the molecular interactions
such as hydrogen bonding between the host and guest molecules
is possible. Although our FT-IR spectra (Fig. S4) did not
conclusively reveal any possible such interaction and while Suh
et al. [23b] have also not performed such an experiment, we
believe that similar interactions as proposed by Suh et al. [23b]
may have also played roles in the differences in adsorption
capacity and release properties among our materials. In another
Samples
Before adsorption (%)
After ibuprofen adsorption (%)
Grafting APTS
Grafting BTSPA
11.8
11.0
18.0
16.2
vinyl groups in VTS and MPTS functionalized samples, respectively,
have unfavorable interaction with ibuprofen, producing less
adsorption capacity for ibuprofen. These results are consistent
with those reported by others for similar functional groups [37].
The % of adsorption of ibuprofen coating on the outer surface of
the materials was obtained by soaking the samples in ibuprofen
solution, then filtering and quickly washing the materials once
with 20 mL ethanol [24g]. The value was found to be 3.6 and 9.8%
for the APTS and BTSPA functionalized samples, respectively. The
corresponding total loading of ibuprofen in 50 mg samples were
30.2 and 33.2 mg, respectively.
´
The rate of drug release of ibuprofen by the samples
functionalized with APTS and BTSPA were found to be essentially
similar (Fig. 8). The diffusional release data for ibuprofen followed
a Fickian diffusion process and it was fitted on the Higuchi
equation (Y ¼ atn) [38–40], where Y is % Release, t is time, and k
report by Vallet-Regı and co-workers [13], not only the functional
groups but also the loading of drug or the drug:host ratio is
reported to also play an important role for drug release kinetics.
Interestingly, the drug release profiles of our samples seem to
show unusually slower drug release followed by faster release.

ARTICLE IN PRESS
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
1659
This kind of release profile is peculiar and, to our best knowledge,
has never been reported for mesoporous materials. However,
we do not know its underlying mechanism, which may warrant
further investigations.
[b] S.P. Rigby, M. Fairhead, C.F. van der Walle, Curr. Pharm. Des. 14 (2008)
1821–1831;
[c] I.I. Slowing, B.G. Trewyn, S. Giri, V.S.-Y. Lin, Adv. Funct. Mater. 17 (2007)
1225–1236;
[d] B.G. Trewyn, I.I. Slowing, S. Giri, H.-T. Chen, V.S.-Y. Lin, Acc. Chem. Res. 40
(2007) 846–853;
[e] F. Balas, M. Manzano, P. Horcajada, M. Vallet-Regı´, J. Am. Chem. Soc. 128
(2006) 8116–8117;
[f] P. Horcajada, A. Ra´mila, G. Ferey, M. Vallet-Regı´, Solid State Sci. 8 (2006)
1243–1249.
4. Conclusion
[5] [a] Z.G. Feng, Y.S. Li, D.C. Niu, L. Li, W. Zhao, H. Chen, L. Li, J. Gao, M. Ruan,
J. Shi, Chem. Commun. (2008) 2629–2631;
[b] A. Sayari, S. Hamoudi, Chem. Mater. 13 (2001) 3151–3168.
[6] [a] A. Stein, B.J. Melde, R.C. Schroden, Adv. Mater. 12 (2000) 1403–1419;
[b] D. Brunel, A.C. Blanc, A. Galarneau, F. Fajula, Catal. Today 73 (2002)
139–152;
The adsorption capacity and release properties of mesoporous
materials for different drug molecules can be improved by
functionalizing them with judiciously chosen organic groups.
Rhodamine 6G and ibuprofen, which have different structures and
surface properties, were used as model drug molecules to conduct
comparative adsorption and release studies of organic functiona-
lized mesoporous materials that were synthesized by post-
grafting and co-condensation. While mesoporous samples func-
tionalized with mercaptopropyl and vinyl groups showed im-
proved adsorption capacity for rhodamine 6G, samples
functionalized with primary and secondary amine groups that
were synthesized by grafting and co-condensation methods
exhibited less adsorption capacity for rhodamine 6G. On the
other hand, the trend in the degree of adsorption of ibuprofen was
found to be the reverse of the results obtained for rhodamine 6G.
The samples containing mercaptopropyl and vinyl groups showed
less adsorption capacity for ibuprofen than the samples functio-
nalized with primary and secondary amine groups. The release of
the adsorbed molecules was also found to be dependent on the
type of functional groups in the materials. The method of tuning
adsorption capacity and release properties by organic functiona-
lization of mesoporous materials could be extended to other
organic functional groups and to a variety of other drug molecules.
The resulting functionalized mesoporous materials may help to
deliver drugs efficiently and, thus, minimize the drugs’ possible
adverse effects. Introducing secondary bioactive groups onto the
external surface of the materials may also allow targeted delivery
of the drug cargo to specific cells.
[c] J. Liu, X.-D. Feng, G.E. Fryxell, L.-Q. Wang, A.-Y. Kim, M.L. Gong, Adv. Mater.
10 (1998) 161–165;
[d] S. Shylesh, A.P. Singh, J. Catal. 244 (2006) 52–64.
[7] R. Mellaerts, J.A.G. Jammaer, M. Van Speybroeck, H. Chen, J. Van Humbeeck, P.
Augustijns, G. Van den Mooter, J.A. Martens, Langmuir 24 (2008) 8651–8659.
[8] [a] S.P. Hudson, R.F. Padera, R. Langer, D.S. Kohane, Biomaterials 29 (2008)
4045–4055;
[b] R. Mellaerts, R. Mols, J.A.G. Jammaer, C.A. Aerts, P. Annaert, J.V. Humbeeck,
G. Van den Mooter, P. Augustijns, J.A. Martens, Eur. J. Pharm. Biopharm. 69
(2008) 223–230.
[9] [a] J.M. Rosenholm, A. Meinander, E. Peuhu, R. Niemi, J.E. Eriksson,
C. Sahlgren, M. Linde´n, ACS Nano 3 (2009) 197–206;
[b] S.R. Blumen, K. Cheng, M.E. Ramos-Nino, D.J. Taatjes, D.J. Weiss,
C.C. Landry, B.T. Mossman, Am. J. Respir. Cell Mol. 36 (2007) 333–342.
´
˜
´
´
[10] [a] A. Ramila, B. Munoz, J. Perez-Pariente, M. Vallet-Regı, J. Sol–gel Sci.
Technol. 26 (2003) 1199–1202;
[b] M. Vallet-Regi, J.C. Doadrio, A.L. Doadrio, I. Izquierdo-Barba, J. Perez-
Pariente, Solid State Ionics 172 (2004) 435–439.
[11] [a] Y.F. Zhu, J.L. Shi, W.H. Shen, X.P. Dong, J.W. Feng, M.L. Ruan, Y.S. Li, Angew.
Chem. Int. Ed. 44 (2005) 5083–5087;
[b] Y.-F. Zhu, J.-L. Shi, Y.-S. Li, H.-R. Chen, W.-H. Shen, X.-P. Dong, J. Mater. Res.
20 (2005) 54–61;
[c] Q. Yang, S.C. Wang, P.W. Fan, L.F. Wang, Y. Di, K.F. Lin, F.S. Xiao, Chem.
Mater. 17 (2005) 5999–6003.
[12] [a] C.D. Nunes, P.D. Vaz, A.C. Fernandes, P. Ferreira, C.C. Roma˜o, M.J. Calhorda,
Eur. J. Pharm. Biopharm. 66 (2007) 357–365;
[b] Q.L. Tang, Y. Xu, D. Wu, Y.H. Sun, J. Solid State Chem. 179 (2006)
1513–1520;
[c] V. Ambrogi, L. Perioli, F. Marmottini, S. Giovagnoli, M. Esposito, C. Rossi,
Eur. J. Pharm. Biopharm. 32 (2007) 216–222.
[13] [a] M. Vallet-Regı´, F. Balas, D. Arcos, Angew. Chem. Int. Ed. 46 (2007)
7548–7558;
[b] R. Mellaerts, C.A. Aerts, J.V. Humbeeck, P. Augustijns, G.V. Mooter,
J.A. Martens, Chem. Commun. 13 (2007) 1375–1377.
[14] F. Qu, G. Zhu, S. Huang, S. Li, J. Sun, D. Zhang, S. Qiu, Microporous Mesoporous
Mater. 92 (2006) 1–9.
Acknowledgments
We gratefully acknowledge the partial financial support by the
US National Science Foundation (NSF), CAREER Grant, CHE-
0645348 for this work. EAB and ANO were partly supported by
NSF-REU (Research Experiences for Undergraduates) program
over Summer 2006 and 2007, respectively.
[15] [a] S.-W. Song, K. Hidajat, S. Kawi, Langmuir 21 (2005) 9568–9575;
´
[b] B. Gonzalez, M. Colilla, M. Vallet-Regi, Chem. Mater. 20 (2008)
4826–4834.
[16] [a] T. Kokubo, H. Kushitani, S. Sakka, T. Kitsugi, T. Yamamuro, J. Biomed.
Mater. Res. 24 (1990) 721–734;
[b] Q. Tang, Y. Xu, D. Wu, Y. Sun, J. Solid State Chem. 179 (2006) 1513–1520.
[17] K.K. Sharma, T. Asefa, Angew Chem. Int. Ed. 46 (2007) 2879–2882.
[18] [a] I.I. Slowing, B.G. Trewyn, S. Giri, V.S.-Y. Lin, Adv. Funct. Mater. 17 (2007)
1225–1236;
Appendix A. Supplementary material
[b] B.G. Trewyn, I.I. Slowing, S. Giri, H.-T. Chen, V.S.-Y. Lin, Acc. Chem. Res. 40
(2007) 846–853;
[c] I.I. Slowing, J.-L. Vivero-Escoto, C.-W. Wu, V.S-Y. Lin, Adv. Drug Delivery
Rev. 60 (2008) 1278–1288;
Supplementary data associated with this article can be found
in the online version at doi:10.1016/j.jssc.2009.03.034.
S. Giri, B.G. Trewyn, V.S.-Y. Lin, Nanomedicine 2 (2007) 99–111.
[d]
[19] [a] F.Y. Qu, G.S. Zhu, S.Y. Huang, S.G. Li, J.Y. Sun, D.L. Zhang, S.L. Qiu,
Microporous Mesoporous Mater. 92 (2006) 1;
[b] H. Vallhov, S. Gabrielsson, M. Stromme, A. Scheynius, A.E. Garcia-Bennett,
Nano Lett. 7 (2007) 3576.
[20] [a] K.K. Sharma, T. Asefa, Angew. Chem. Int. Ed. 46 (2007) 2879–2882;
[b] K. Moller, T. Bein, Chem. Mater. 10 (1998) 2950–2963;
[c] G.E. Fryxell, J. Liu, T.A. Hauser, Z. Nie, K.F. Ferris, S. Mattigod, G. Meiling,
R.T. Hallen, Chem. Mater. 11 (1999) 2148–2154.
[21] [a] J.A. Melero, R. van Grieken, G. Morales, Chem. Rev. 106 (2006) 3790–3812;
[b] M.H. Lim, C.F. Blanford, A. Stein, J. Am. Chem. Soc. 119 (1997) 4090–4091;
[c] M. Kruk, T. Asefa, M. Jaroniec, G.A. Ozin, Stud. Surf. Sci. Catal. 141 (2002)
197–204;
References
[1] [a] M. De, P.S. Ghosh, V.M. Rotello, Adv. Mater. 20 (2008) 1–17;
[b] P. Kolhe, E. Misra, R.M. Kannan, S. Kannan, M. Lieh-Lai, Int. J. Pharm. 259
(2003) 143–160.
[2] [a] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359
(1992) 710–712;
[b] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt,
C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins,
J.C. Schlenker, J. Am. Chem. Soc. 114 (1992) 10834–10843.
[3] [a] T. Yanagisawa, T. Shimizu, K. Kuroda, C. Kato, Bull. Chem. Soc. Jpn. 63
(1990) 1535–1537;
[d] I. Slowing, B.G. Trewyn, V.S.-Y. Lin, J. Am. Chem. Soc. 128 (2006)
14792–14793.
[22] S. Huh, J.W. Wiench, J.-C. Yoo, M. Pruski, V.S.-Y. Lin, Chem. Mater. 15 (2003)
4247–4256.
[b] S. Inagaki, Y. Fukushima, K. Kuroda, J. Chem. Soc., Chem. Commun. (1993)
680–682.
[23] [a] Y. Cho, R. Shi, R.B. Borgens, A. Ivanisevic, Nanomedicine
507–519;
3 (2008)
[4] [a] M. Vallet-Regi, A. Ra´mila, R.P. del Real, J. Pe´rez-Pariente, Chem. Mater. 13
(2001) 308–311;
[b] M. Suh, H.-J. Lee, J.-Y. Park, U.-H. Lee, Y.-U. Kwon, D.J. Kim, Chem. Phys.
Chem. 9 (2008) 1402–1408;

ARTICLE IN PRESS
1660
G. Wang et al. / Journal of Solid State Chemistry 182 (2009) 1649–1660
´
´
´
[c] B. Munoz, A. Ramila, J. Perez-Pariente, I. Diaz, M. Vallet-Regı, Chem. Mater.
[28] A. Mehdi, C. Reye, S. Brandes, R. Guilard, R.J.P. Corriu, New J. Chem. 29 (2005)
15 (2003) 500–503;
965–968.
[d] S.-W. Song, K. Hidajat, S. Kawi, Langmuir 21 (2005) 9568–9957;
[e] I. Izquierdo-Barba, A. Martinez, A.L. Doadrio, J. Pe´rez-Pariente, M. Vallet-
Regı´, Eur. J. Pharm. Biopharm. 26 (2005) 365–373;
[f] T. Sekine, K. Nakatani, Langmuir 18 (2002) 694–697;
[g] J.C. Doadrio, E.M.B. Sousa, I. Izquierdo-Barba, A.L. Doadrio, J. Perez-
[29] F. Qu, G. Zhu, S. Huang, S. Li, J. Sun, D. Zhang, S. Qiu, Microporous Mesoporous
Mater. 92 (2006) 1.
[30] H. Vallhov, S. Gabrielsson, M. Stromme, A. Scheynius, A.E. Garcia-Bennett,
Nano Lett. 7 (2007) 3576–3582.
[31] [a] M. Kruk, M. Jaroniec, Chem. Mater. 13 (2001) 3169–3183;
[b] K. Fodor, J.H. Bitter, K.P. de Jong, Microporous Mesoporous Mater. 56
(2002) 101–109;
[c] M. Kruk, M. Jaroniec, Chem. Mater. 12 (2000) 222–230.
[32] [a] M. Kruk, T. Asefa, M. Jaroniec, G.A. Ozin, J. Mater. Chem. 12 (2002)
3452–3457;
´
Pariente, M. Vallet-Regı, J. Mater. Chem. 16 (2006) 462–466.
[24] [a] K. Patel, S. Angelos, W.R. Dichtel, A. Coskun, Y.-W. Yang, J.I. Zink,
J.F. Stoddart, J. Am. Chem. Soc. 130 (2008) 2382–2383;
[b] F. Qu, G. Zhu, H. Lin, J. Sun, D. Zhang, S. Li, S. Qiu, Eur. J. Inorg. Chem.
(2006) 3943–3947;
[c] P. Yang, S. Huang, D. Kong, J. Lin, H. Fu, Inorg. Chem. 46 (2007) 3203–3211;
[d] M. Vallet-Regı´, A. Ramila, R.P. Real, J. Perez-Pariente, Chem. Mater. 13
(2001) 308–311;
[e] Y.F. Zhu, J.L. Shi, Y.S. Li, H.R. Chen, W.H. Shen, X.P. Dong, Microporous
Mesoporous Mater. 85 (2005) 75–81;
[b] L. Pasqua, F. Testa, R. Aiello, S. Cundari, J.B. Nagy, Microporous
Mesoporous Mater. 103 (2007) 166–173.
[33] M.H. Lim, C.F. Blanford, A. Stein, Chem. Mater. 10 (1998) 467–470.
[34] [a] S. Zhu, Z. Zhou, D. Zhang, C. Jin, Z. Li, Microporous Mesoporous Mater. 106
(2007) 56–61;
[f] T. Heikkila¨, J. Salonen, J. Tuura, M.S. Hamdy, G. Mul, N. Kumar, T. Salmi, D.Y.
Murzin, L. Laitinen, A.M. Kaukonen, Int. J. Pharm. 331 (2007) 133–138;
[b] C.-Y. Hong, X. Li, C.-Y. Pan, J. Phys. Chem. C 112 (2008) 15320–15324;
[c] X. Yu, S. Ding, Z. Meng, Z. Yang, Colloid Polym. Sci. 286 (2008) 1361–1368;
[d] L. Zhang, J. Liu, J. Yang, Q. Yang, C. Li, Microporous Mesoporous Mater. 109
(2008) 172–183.
´
´
´
[g] C. Charnay, S. Begu, C. Tourne-Peteilh, L. Nicole, D.A. Lerner, J.M.
Devoisselle, Eur. J. Pharm. Biopharm. 57 (2004) 533–540.
[25] [a] U. Brohede, R. Atluri, A.E. Garcia-Bennett, M. Stromme, Curr. Drug
Delivery 5 (2008) 177–185;
[35] M.V. Cagnoli, S.G. Casuscelli, A.M. Alvarez, J.F. Bengoa, N.G. Gallegos,
M.E. Crivello, E.R. Herrero, S.G. Marchetti, Catal. Today 107–108 (2005)
397–403.
[36] A. Walcarius, M. Etienne, B. Lebeau, Chem. Mater. 15 (2003) 2161–2173.
[37] F. Qu, G. Zhu, H. Lin, J. Sun, D. Zhang, S. Li, S. Qiu, Eur. J. Inorg. Chem. 19 (2006)
3943–3947.
[b] Y.-Z. You, K.K. Kalebaila, S.L. Brock, D. Oupicky, Chem. Mater. 20 (2008)
3354–3359;
[c] H.Y. Huang, C.L. Yang, H. Zhang, M. Liu, Microporous Mesoporous Mater.
111 (2008) 254–259;
[d] S. Shen, P.S. Chow, S. Pui, S. Kim, K. Zhu, R.B.H. Tan, J. Colloid Interface Sci.
321 (2008) 365–372;
[e] J. Lu, E. Choi, F. Tamanoi, J.L. Zink, Small 4 (2008) 421–426.
[26] S. Sadhasivam, S. Savitha, K. Swaminathan, J. Environ. Manag. 85 (2007) 155–161.
[38] P. Costa, J.M.S. Lobo, Eur. J. Pharm. 13 (2001) 123–133.
[39] T. Heikkila¨, J. Salonen, J. Tuura, M.S. Hamdy, G. Mul, N. Kumar, T. Salmi, D.Y.
Murzin, L. Laitinen, A.M. Kaukonen, J. Hirvonen, V.-P. Lehto, Int. J. Pharm. 331
(2007) 133–138.
[27] K.D. Rainsford, Ibuprofen,
Philadelphia, PA, 1999.
A
Critical Biographical Review, CRC Press,
[40] J. Andersson, J. Rosenholm, S. Areva, M. Linde´n, Chem. Mater. 16 (2004)
4160–4167.