1. Adsorption characteristics of a cationic porphyrin on nanoclay at various pH
Zachary Rice, Magnus Bergkvist *
University at Albany, State University of New York, College of Nanoscale Science and Engineering, Albany, NY 12203, USA
a r t i c l e i n f o
Article history:
Received 14 November 2008
Accepted 22 February 2009
Available online 10 April 2009
Keywords:
Montmorillonite
Nanoclay
Porphyrin
Adsorption
a b s t r a c t
Natural and synthetic porphyrin derivatives offer a range of applications including enzymatic catalysis,
photosensitizers for light harvesting and chemical reactions, and molecular electronics. They exhibit
unique optical spectra dominated by the presence of Soret and Q-band structures whose position and
shape offer a straightforward method to characterize porphyrins in various surroundings. In many appli-
cations it is often beneficial to have porphyrins adsorbed onto a solid matrix. Applications of porphyrin–
clay complexes extend to numerous biological applications including pharmaceutical drug delivery, cos-
metics, and agricultural applications and thus a full understanding of porphyrin–clay surface interactions
are essential. Here we investigated the adsorption behavior of meso-tetra(4-N,N,N-trimethylanilinium)
porphine (TMAP) onto sodium containing, natural montmorillonite clay (Cloisite Na+
) in characteristic
biological buffers over a range of pHs ($2–9). Spectroscopic analyses show a linear absorption response
at acidic and basic pHs but a slight deviation at intermediate pHs. Absorption spectra for TMAP on clay
showed distinct red shifts of the Soret and Q-bands compared to free TMAP for all buffer conditions indi-
cating core p-electron delocalization into the substituent rings. At intermediate pHs, a gradual transition
between protonated/deprotonated states were seen, presumably due to higher H+
concentration at the
surface than in bulk. Results indicate TMAP adsorption to clay occurs in a monolayer fashion at low/high
pH while slightly acidic/neutral pH possibly rearrange on the surface and/or form aggregates. AFM
images of clay saturated with TMAP are reported and show single isolated clay sheets without aggrega-
tion, similar to clay without TMAP.
Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction
Natural and synthetic porphyrin derivatives are heterocyclic
macrocycle compounds with interesting optical and chemical
properties. In biological systems they can be found in various pho-
tosynthetic light harvesting complexes, enzymes, and transport
proteins. In addition, porphyrins are of interest for a range of appli-
cations including enzymatic catalysis [1], photosensitizers for light
harvesting and chemical reactions [2–5] and molecular electronics
[6]. Depending on the application porphyrins are either used dis-
solved in bulk solution or adsorbed to solid substrates. Owing to
their unique structure, they exhibit distinctive optical spectra,
dominated by the presence of the so-called Soret band which is
caused by an electron transition from the ground-state singlet to
the second excited singlet state, and a Q-band structure caused
by a ground-state singlet transition to the first excited singlet state
[7]. The position and shape of these absorption bands are influ-
enced by changes in the molecular structure as well as environ-
mental effects and offer a straightforward method to characterize
porphyrins in various surroundings using UV–vis and fluorescent
spectroscopy. For example, bathochromic shifts (red shifts) in the
spectra can be indicative of porphyrin side-by-side stacking in
solution (J-aggregates) causing a electron delocalization, or it can
be caused by flattening of the molecule upon adsorption to a sur-
face in where phenyl substituents are rotated in relation to the
main porphine ring (Fig. 1) [3,8,9]. In this case, the red shift stems
from a delocalization of p-electrons throughout the molecule via
the electron withdrawing phenyl groups [9–11].
In many practical applications it is often beneficial to have por-
phyrins adsorbed onto a solid matrix. In this regard, nanoparticles
or ‘‘nanoclay” minerals offer unique opportunities providing well-
characterized interfaces and a high surface area. Nanoclay is a
loose term but is often used for a group of expanding clay minerals
known as smectites including hectorite, saponite, beidelite, nontr-
onite, and montmorillonite. Various smectite nanoclay materials
have been utilized in a range of areas such as light harvesting
[12], surface chemistry applications [12,13], polymer blends
[14,15], and as drug delivery vehicles [16]. For this study we used
a natural montmorillonite, which is the main constituent of the
clay bentonite and possess unique hydration, swelling, and adsorp-
tion properties. It is an alumina silicate mineral containing a single
alumina octahedral layer sandwiched between two hexagonally
0021-9797/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2009.02.070
* Corresponding author. Address. University at Albany, State University of New
York, College of Nanoscale Science and Engineering, 257 Fuller Road, Albany, NY
12203, USA.
E-mail addresses: MBergkvist@uamail.albany.edu, magnus.bergkvist@gmail.com
(M. Bergkvist).
Journal of Colloid and Interface Science 335 (2009) 189–195
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science
www.elsevier.com/locate/jcis

2. close-packed silica tetrahedral double layers (Fig. 2) [13]. These
clay sheets have a high negative surface charge and are naturally
found stacked on top of each other, separated by positively charged
cations. Upon exfoliation, larger clay particles can separate to give
individual 100–400 nm wide sheets of alumina silicate sandwich
layers, resulting in ‘‘nanoclay” sheets (Fig. 3a). The montmorillon-
ite clays typically have a high cation exchange capacity (CEC)
where positively charged molecules (such as cationic porphyrins)
can either be substituted into the interlamellar space between
stacked sheets or adsorbed to the surface of single exfoliated
sheets.
Studies of the adsorption and interaction of various porphyrins
with solid interfaces, such as the synthetic clay Sumecton SA, have
revealed various shifts in porphyrin UV–vis absorption spectra
[3,4,9,17], interclay spacing distance [3,18,19], and porphyrin ring
distortion [9,20–22] illustrating the multifaceted surface interac-
tions of these molecules (see Takagi et al. for a recent review
[2]). For the most part studies have been performed in pure dis-
tilled water; however as porphyrin–nanoclay constructs are attrac-
tive in various nanotechnology and biology applications their
adsorption behavior in buffered aqueous solutions at various pHs
is of interest.
Here we investigated the adsorption behavior of meso-tetra(4-
N,N,N-trimethylanilinium) porphine (TMAP) (Fig. 1) onto sodium
containing, natural montmorillonite clay (Cloisite Na+
) in charac-
teristic biological buffers over a range of pHs. Various loading con-
centrations of TMAP vs. CEC were used where TMAP adsorption
behavior was monitored by UV–vis and fluorescence spectroscopy.
Clay sheet morphology was investigated using atomic force
microscopy (AFM). Results indicated that TMAP molecules ad-
sorbed on the nanoclay surfaces in a monolayer fashion for both
high/low pH. A red shift of the Soret band indicate a flattening of
the ring substituent groups upon adsorption and a pH shift of the
ring protonation compared to bulk solutions, presumably due to
accumulation of protons at the surface. All loading concentrations
result in an appreciable distance between individual clay sheets in
solution and show no evidence of close-packed montmorillonite
layer stacking. Upon dehydrating the sample or freeze–thawing,
a close packed stacking can be induced.
Understanding the effects of various pH buffers on the UV–Vis
spectrum allows a better understanding of how these porphyrin–
clay complexes stabilize under a variety of conditions. Applications
of porphyrin–clay complexes extend to numerous biological appli-
cations including pharmaceutical drug delivery, cosmetics, and
agricultural applications [23] and thus a full understanding of por-
phyrin–clay surface interactions are essential.
2. Methods
2.1. Chemicals
Cloisite Na+
, a natural form of montmorillite, was obtained from
Southern Clay Products, Inc., Gonzalez, TX and used as received.
Cloisite clay has a cation exchange capacity (CEC) of 92–93 meq/
100 g and a surface area of 750 m2
/g as determined by the supplier.
Meso-tetra(4-N,N,N-trimethylanilinium) porphine tetrachloride
(TMAP - # T973) was obtained from Frontier Scientific (Logan,
UT). Hydrochloric acid (HCl) 99.5% purity, Boric Acid (Borate)
99.5% purity, and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic
acid (HEPES) 99% purity were obtained from Sigma Chemicals Co.
(St. Louis, MO). 2-(N-morpholino)ethanesulfonic acid (MES) 99%
purity was obtained from TEKnova (Hollister, CA). 18.3 MX cm
deionized water (dH2O) was prepared using a Millipore Biocell
System. All chemicals were used as received.
2.2. Clay-buffer preparation
Cloisite clay was added to various 10 mM buffer solutions to
give stock solutions of 0.1 mg/ml as follows, HCl (pH 1.75), MES
(pH 4.5), HEPES (pH 7.2), and Borate (pH 9.0). In addition a
0.1 mg/ml solution of clay in dH2O was also prepared. All solutions
were then magnetically stirred at 500 RPM for at least 10 h to exfo-
liate the clay. The reported pH values were measured after mixing/
exfoliation of the clay.
2.3. Porphyrin adsorption on clay
A 50 lM TMAP stock solution was prepared in dH2O (based on a
absorption coefficient of 2.42 Â 105
MÀ1
cmÀ1
at 412 nm). This
TMAP concentration would give >200% theoretical loading of por-
phyrin on clay if mixed in a 1:1 volume ratio with our clay-buffer
solutions. Calculations are based on CEC of 93 meq/100 g and a
total +4 charge on TMAP. 100% loading vs. CEC means that all neg-
atively charged sites on the clay are associated with TMAP mole-
cules, i.e. one TMAP molecule ‘‘neutralizes” 4 charged sites on
the clay. A fixed volume of each clay stock solution was mixed with
Fig. 2. Schematic structure of montmorillonite clay. Illustration generated with the
GDIS software (released under the GNU General Public License) using crystal
coordinates from the American Mineralogist Crystal Structure Database.
N
+
N
+
N
+
NH
N
NH
N
N
+
Cl
-
Cl
-
Cl
-
Cl
-
Fig. 1. Molecular representation of meso-tetra(4-N,N,N-trimethylanilinium) por-
phine tetrachloride (TMAP). Substituent groups on the porphyrin core can rotate
freely and are in solution essentially perpendicular to the plane of the core.
190 Z. Rice, M. Bergkvist / Journal of Colloid and Interface Science 335 (2009) 189–195

3. an equal volume made up of various ratios of TMAP stock and cor-
responding buffer to give 0–150% final loading vs. CEC. In addition,
equivalent TMAP solutions without any clay added were prepared
for each loading concentration and pH.
2.4. Spectroscopic analysis
Absorbance spectra were obtained using a TECAN Infinite M200
series spectrometer scanning from 230 to 750 nm with 1 nm incre-
ments. The same instrument was used to obtain fluorescent spec-
tra. Excitation was at 428 nm and the emission was scanned
between 500 and 800 nm for all samples. Data were plotted and
analyzed using Origin (OriginLab, MA).
2.5. AFM analysis
Samples were prepared by using freshly cleaved muscovite
mica sheets as a substrate. Clay–TMAP (5 ll) particles prepared
as above were deposited onto the mica surface and left to settle
for 5 min and dried with N2 gas. AFM measurements were per-
formed with a Nanoscope III Multimode AFM in tapping mode
(Digital Instruments, CA) using standard AFM silicon tips
(300 kHz, 40 N/m) (Olympus) resonated at a free amplitude of
$2.5 V. Images were recorded with 512 Â 512 pixel resolution
at 1 Hz scan rate.
2.6. Zeta potential measurements
Zeta potential measurements were conducted in a Zetasizer ZS
instrument (Malvern Instruments, MA) employing laser Doppler
velocimetry and phase analysis light scattering. Zeta potentials
were calculated via the instrument software based on the Henry
equation and Smoluchowski model approximation.
3. Results and discussion
3.1. Absorption spectroscopy – TMAP in solution
Absorbance spectra of free TMAP in various pH buffers can be
seen in Fig. 4. Nearly all buffer solution spectra had Soret bands
Fig. 3. AFM images of (a) native Cloisite Na+
in pH 9.0 buffer, (b) Cloisite clay at 100% TMAP loading vs. CEC (pH 9.0). These images show that TMAP adsorbs onto clay sheets
without significant stacking/aggregation of the sheets. Although clay sheets sediment over time they are readily dispersed by vortexing (with or without TMAP).
360 380 400 420 440 460 480 500
0.0
0.1
0.2
0.3
0.4
0.5
0.6
500 520 540 560 580 600 620 640 660 680 700
0.00
0.01
0.02
0.03
0.04
0.05
432
a.u
nm
TMAP in 10mM HCl (pH1.75)
TMAP in 10mM MES (pH 4.5)
TMAP in 10mM HEPES (pH 7.2)
TMAP in 10mM Borate (pH 9.0)
TMAP in water
413
TMAP in solution (no clay)
590
640
632
552 580515
Fig. 4. Absorption spectra of free TMAP in various pH buffer solutions. For pH P 4.5 kmax is at 413 nm. At low pH protonation of the porphine ring shifts the Soret peak to
432 nm. Insert show the Q-bands where upon protonation the ring symmetry shift from D2h to D4h, reducing the number of peaks and their position (four peak to two peaks).
Z. Rice, M. Bergkvist / Journal of Colloid and Interface Science 335 (2009) 189–195 191

4. with kmax at 413 nm and Q-bands present at 515, 552, 580, and
632 nm (corresponding to a D2h symmetry). For TMAP in 10 mM
HCl a bathochromic red shift of $20 nm placed the kmax at
432 nm. Likewise, the Q-bands red shifted and the intensity profile
changed from a four peak to a two peak spectra indicative of D4h
symmetry in where both pyrollic nitrogens are protonated. The
pKa values for mono/di-protonation of the core nitrogens are 4.11
and 3.95, respectively, hence the characteristic bathochromic and
intensity profile shifts observed at pH 1.75 [19].
3.2. Absorption spectroscopy – TMAP on clay
Absorption spectra for TMAP on clay showed distinct red shifts
compared to free TMAP for all buffer conditions. However there
were differences in peak positions among samples depending on
buffer condition and loading (see Fig. 5a–d for peak values). TMAP
loading studies on nanoclay were performed by measuring the
amount of TMAP remaining in the supernatant after a centrifuga-
tion step (16,000 RCF) for each of the loading concentrations. There
was no trace of TMAP in the supernatants for any of the buffers be-
low 100% loading. Above 100% loading vs. CEC we noticed gradu-
ally increasing amount of ‘‘free” TMAP in the supernatant. This
indicated that available binding sites on the clay sheets were occu-
pied and that approximately one TMAP molecule bound per four
surface charges. The exact transition point varied slightly between
the various sample buffers but all were between 95% and 115% vs.
CEC (higher transition for higher pH). Resuspension of the clay pel-
let in fresh buffer followed by an additional centrifugation showed
no trace of desorbed TMAP indicating a reasonably strong binding.
It can be noted that all TMAP/clay solutions produced stable sus-
pensions for TMAP loadings up to 95%. These suspensions slowly
flocculated over time (hours) but were easily redispersed by agita-
tion. For higher loading concentrations the flocculation was faster
(min–h); however clay particles were still readily re-suspended.
AFM images of clay with $100% loading showed single isolated
clay sheets without aggregation, similar to clay without TMAP
(Fig. 3b).
3.2.1. Low/high pH buffers
Absorption spectra of samples prepared in borate (pH 9.0) had
similar intensity profiles as free TMAP but with red shifted Soret
and Q-bands. The kmax at 426 nm represent a $14 nm shift. At high
loading levels (>100%) there was an increasing signal from free
TMAP and at 150% loading, free TMAP cause a apparent broadening
of the Soret region (Fig. 5d). Samples prepared in HCl (pH 1.75) had
characteristic spectra of protonated TMAP but with the Soret and
Q-band red shifted. The kmax at 436 nm correspond to a shift of
4 nm compared to free TMAP (Fig. 5a). For the pH 1.75 samples
the Soret peaks from free and clay-bound TMAP overlapped signif-
icantly making it hard to distinguish individual peaks; however
(d)
(c)
(b)
360 380 400 420 440 460 480 500
0.0
0.5
1.0
500 520 540 560 580 600 620 640 660 680 700
0.0
0.1
427
a.u
nm
30% load
60% load
95% load
150% load
10 mM HEPES, pH 7.2
655
525
565
658
360 380 400 420 440 460 480 500
0.0
0.5
1.0
1.5
500 520 540 560 580 600 620 640 660 680 700
0.0
0.1
426
10 mM Borate, pH 9.0
a.u
nm
30% load
60% load
95% load
150% load
655595565525
360 380 400 420 440 460 480 500
0.0
0.5
1.0
1.5
500 520 540 560 580 600 620 640 660 680 700
0.0
0.1
436
413
a.u
nm
30% load
60% load
95% load
150% load
10 mM MES, pH 4.5
521
595
644
360 380 400 420 440 460 480 500
0.0
0.5
1.0
1.5
2.0
2.5
500 520 540 560 580 600 620 640 660 680 700
0.0
0.1
0.2
0.3
436
a.u
nm
30% load
60% load
95% load
150% load
10 mM HCl, pH 1.75
644
595
(a)
Fig. 5. Absorption spectra for clay with various loading of TMAP where (a) 10 mM HCl pH 1.75, (b) 10 mM MES pH 4.5, (c) 10 mM HEPES pH 7.2, and (d) 10 mM Borate pH 9.0.
The red shift of TMAP adsorbed to clay in comparison to free TMAP at equivalent pH is assigned to a flattening of TMAP ring substituents and a delocalization of p-electrons.
Q-bands are inserted and major absorption peaks are indicated. Spectra of TMAP–clay at pH 4.5 indicate ring protonation (not seen in for free TMAP at this pH).
192 Z. Rice, M. Bergkvist / Journal of Colloid and Interface Science 335 (2009) 189–195

5. contribution from free TMAP at loading levels above 100% was evi-
denced as a slight blue shift of the Soret peak. The spectral red
shifts observed for the TMAP/clay samples could arise from phe-
nomena such as porphyrin stacking/aggregation or substituent
group rotation. It has been shown that porphyrin adsorption to
various surfaces; including synthetic smectite clays, induce a rota-
tion of the phenyl substituents from a perpendicular to a more par-
allel orientation related to the porphine ring. This leads to a
‘‘flattening” of the molecule where the core p-electrons can be fur-
ther delocalized into the substituent rings, causing a red shift of the
Soret and Q-bands [9,10,17,24]. The combination of the loading vs.
CEC results and the observed red shifts indicated that individual
TMAP molecules adsorbed without aggregation onto natural mont-
morillite clay in a ‘‘flattened” orientation at low, and high pH. The
main distinction between the HCl and borate buffers was that
TMAP most likely adsorbed in a di-protonated form at a low pH.
The exact cause of the smaller Soret band shift (4 nm) for samples
at pH 1.75 compared to the larger shift for basic pH (14 nm) is un-
clear at this point. It could result from reduced p-electron delocal-
ization as an effect of the protonated core, different amount of
rotation for the phenyl groups upon adsorption at the various
pH’s or a combination of factors.
3.2.2. Intermediate pH buffers
Compared to above results, TMAP adsorbed to cloisite in slightly
acidic (MES, pH 4.5) or neutral (HEPES, pH 7.2) buffer conditions
exhibit more complex absorption spectra. In HEPES, the Soret peak
had comparable red shift as for borate and at first glance the Q-
bands also looked similar. This indicated a similar adsorption as
for pH 9.0. However, upon closer inspection the third Q-band peak
appeared ‘‘compressed” and was slightly red shifted (Fig. 5c).
When TMAP loadings were >100% vs. CEC a more ‘‘typical” Q-band
profile appeared due to free TMAP (Fig. 5c). Adsorption of porphy-
rin to clay in MES buffer demonstrated unique spectra characteris-
tics. At lower loading concentrations the Soret band had a kmax at
436 nm which indicated that TMAP adsorbed in a protonated state
analogous to TMAP/clay in HCl. Likewise the Q-band profiles for
these loadings corresponded to a protonated core (Fig. 5b). In com-
parison, free TMAP molecules in MES buffer were deprotonated
(Fig. 4). At loading levels between 60% and 95% the Q-bands took
on characteristics between the clay/HCl and clay/HEPES samples,
i.e. a substantial peak at $644 nm but with increased peak inten-
sity at 565 and $525 nm (Fig. 5b). Also, the Soret region became
wider indicating a transition. This widening could be interpreted
as a contribution from free TMAP in solution; however our centri-
fugation/supernatant results showed no trace of free TMAP below
100% loading. Exfoliation of clays in water leave the surface highly
negatively charged and to counter this charge positive ions accu-
mulate at the interface. As a consequence, especially in low ionic
strength conditions, the local concentration of H+
is higher than
in bulk which changes the pH near the surface. At low/slightly
acidic bulk pH more protons are available to counter the surface
charge and TMAP molecules at the clay interface encounter a more
acidic environment. Charged TMAP can hypothetically replace the
H+
near the surface as the loading increases which could affect the
surface pH. MES buffer had pH 4.5, which is close to the pKa of the
porphyrin core ($4) and it may well be that the effective surface
pH is below this value. Therefore TMAP in slightly acidic pH could
adsorb on clay in a protonated state at lower loading concentra-
tions, mimicking the results at pH 1.75. Increased surface concen-
tration of TMAP shifted the spectra toward that of TMAP adsorbed
at higher pH. This is indicative of a change in the adsorbed state of
the molecule, possibly associated with deprotonation or a rear-
rangement of TMAP on the surface. Soret band transitions ascribed
to surface proton concentration have been observed for a pyridyl
substituted porphyrin adsorbed to different clays in water [25].
In HEPES buffer no effects of loading concentration could be seen
and the results point to a deprotonated core, i.e. the surface pH
is most likely above the porphyrin pKa. However some minor
change in the adsorbed state seems likely when compared to pH
9.0 as the Q-band profile is different.
Additional spectroscopic analyses were conducted to investi-
gate the adsorption behavior of TMAP. Molecular interaction and
energy resonance between porphyrin molecules on surfaces can
cause a deviation from Lambert–Beer’s law and affect fluorescent
properties (reduced emission, shorter excited state lifetimes etc.)
[3,26]. For our samples the graphs of TMAP loading vs. absorption
(kmax) showed good linearity for acidic and basic pHs but deviated
slightly for the intermediate pHs with r2
values below 0.99 (Fig. 6).
We attribute the decreased r2
values to the lower than expected
absorption at higher (95%) loading concentration. Typical fluores-
cent emission spectra were observed for all samples over the
whole TMAP loading range. Emission spectra were similar to TMAP
adsorbed to other surfaces (Fig. 7) [3]. Samples prepared in pH 7.2
appeared similar to pH 9.0 samples (Fig. 7c and d) with emission
max at $675 nm while lower pH solution had maxima at
663 nm. Samples prepared in pH 4.5 showed somewhat more com-
plex emission spectra in where peak profile change at $95% load-
ing (Fig. 7b).
Absorption and fluorescent spectra indicate that TMAP ad-
sorbed to clay surfaces at pH 4.5 and pH 7.2 do so in a different
manner than for higher/lower pH values. At high and low pH,
TMAP likely adsorbs as individual molecules in a flat orientation
very similar as observed for Sumecton clays [3]. In this case, the
substituent groups are flattened out causing the red shift of the
Soret and Q-bands where at low pH the core is protonated. Sam-
ples prepared at pH 4.5 appear to deviate most where TMAP mol-
ecules appeared to change conformation at higher loading
concentrations. Our speculation is that the proximity of the bulk/
surface pH to the pKa of the TMAP might promote intra-molecular
interaction and reorientation. At lower loading the molecules
seemed to adsorb in a flat orientation with the core protonated
as a net effect of the relative higher surface pH. Increased surface
loading could have minute influence on the local surface environ-
ment causing molecules to reorient, leading to partial orbital over-
lap and changed spectral properties.
As a last test we performed zetapotential measurements for
clay at 0% and 150% loadings (Table 1). All clays with 0% loading
0 20 40 60 80 100 120
0.0
0.5
1.0
1.5
R
2
= 0.992
R
2
= 0.997
R
2
= 0.967
R
2
= 0.957
pH 1.75
pH 4.5
pH 7.2
pH 9.0
a.u
TMAP loading vs CEC (%)
Fig. 6. Graphs of kmax intensity of TMAP on clay at various loading concentration
against CEC (436 nm for pH 1.75/4.5 and 426 nm.for pH 7.2/9.0). Linear fits and R2
values for each solution is shown in figure. Intermediate pH buffers tend do deviate
form a linear response at higher TMAP loading concentration.
Z. Rice, M. Bergkvist / Journal of Colloid and Interface Science 335 (2009) 189–195 193

6. had increasing negative zetapotential with pH as expected. Unfor-
tunately samples prepared at pH 1.75 did not give reliable data,
most likely due to corrosion of the electrodes in the sample cuv-
ette. At 150% TMAP loading both pH 9.0 and pH 7.2 samples had
negative surface potential albeit less than for clay alone. This indi-
cated that even though the surface is presumably filled with TMAP
(according to our initial loading studies), some negative sites might
be unoccupied, some TMAP can be ‘‘pulled off” the surface by the
electric field, or that additional OHÀ
associate with the charged
TMAP molecules adsorbed at higher pH. The most interesting
observation however was that at pH 4.5 clay sheets had a positive
surface potential. These results indicated that TMAP was strongly
adsorbed to the surface and caused a charge reversal. The cause
of this charge reversal can be due to a variety of factors. A massive
accumulation/aggregation of TMAP on the surface seems an unli-
kely explanation for this, as our loading studies showed free TMAP
in solution >100% loading. Charge reversal could be a result of
reorientation of protonated TMAP (with a +6 total charge) at high
loading in where molecules are partly bound and do not lie flat
on the surface. This might also be the reason for the atypical
absorption and emission spectra for samples at pH 4.5. Further
studies are required to fully elucidate observed phenomena.
Finally, to compare our results with previous work of porphy-
rins adsorbed to synthetic clays we performed our UV–vis absorp-
tion experiments in dH2O without buffer and pH adjustment
(starting pH of dH2O was $6.5). Upon adding the clay to dH2O
the pH shifted to $9.5, indicating the tendency of clays to ‘‘leach”
protons from low ionic strength solutions. The results in water
were virtually identical to our borate buffer experiments (pH
9.0). Interestingly TMAP adsorption to clay in borate buffer had
the closest resemblance to previous porphyrin/clay work per-
formed in distilled water, indicating that porphyrin adsorption in
these cases most likely occurred at raised pH.
4. Conclusion
This work investigated the adsorption behavior of TMAP to nat-
ural montmorillite nanoclay. Adsorption of TMAP to clay at various
loading percentages was followed using spectroscopy techniques
in buffers ranging from pH $2 to pH 9. Upon adsorption to clay
there was a characteristic bathochromic red shift of the spectra
compared to TMAP in solution. This shift was observed for all pH
conditions, however, at lower pH the TMAP adsorbed in a proton-
ated state as evidenced by the Q-band profile and Soret peak loca-
tion. The red shift was most likely due to p-electron delocalization
upon rotation of substituent groups when TMAP adsorbed to the
clay. At low and high pH conditions, results indicated that TMAP
adsorbed in a monolayer fashion parallel to the clay surface for
all loading conditions. For intermediate pHs, a gradual transition
between protonated/deprotonated states were seen, presumably
due to higher H+
concentration at the surface than in bulk. Also,
spectra showed a loading dependent adsorption behavior at inter-
mediate pHs possibly indicating a re-orientation/interaction of
TMAP molecules on the surface. These observations complement
other previous clay–porphyrin systems studies. Applications of
porphyrin–clay complexes extend to numerous biological applica-
tions including pharmaceutical drug delivery, cosmetics, and agri-
cultural applications, hence understanding the influence of pH/
buffer conditions on porphyrin/clay interactions is of importance.
600 650 700 750 800
RFU
nm
30% load pH 9.0
60% load pH 9.0
95% load pH 9.0
150% load pH 9.0
TMAP pH 9.0
675
652
703
600 650 700 750 800
RFU
nm
30% load pH 7.2
60% load pH 7.2
95% load pH 7.2
150% load pH 7.2
TMAP pH 7.2
676
649
703
600 650 700 750 800
RFU
nm
(b)
(c)
(d)
30% load pH 4.5
60% load pH 4.5
95% load pH 4.5
150% load pH 4.5
TMAP pH 4.5
663
703
649
600 650 700 750 800
663
RFU
nm
(a)
30% load pH 1.75
60% load pH 1.75
95% load pH 1.75
150% load pH 1.75
TMAP in pH 1.75
669
Fig. 7. Emission spectra for clay with various loading of TMAP where (a) 10 mM HCl
pH 1.75, (b) 10 mM MES pH 4.5, (c) 10 mM HEPES pH 7.2, and (d) 10 mM Borate pH
9.0. Excitation at 428 nm for all solutions. Emission peaks for adsorbed TMAP shifts
for all pHs when compared to free TMAP (gray dotted curve). The peak profiles
change at low pH as the porphyrin ring gets protonated.
Table 1
Zeta potential (mV) of Cloisite clay sheets in different pH buffer and TMAP loading. No
stable measurement conditions could be realized at pH 1.75, thus no values are
reported.
Buffer solution 0% TMAP loading 150% loading
HCl (pH 1.75) – –
MES (pH 4.5) À18.5 ± 0.4 38.0 ± 1.0
HEPES (pH 7.2) À36.8 ± 1.3 À7.3 ± 1.1
Borate (pH 9.0) À39.1 ± 0.6 À20.0 ± 1.8
194 Z. Rice, M. Bergkvist / Journal of Colloid and Interface Science 335 (2009) 189–195

7. Acknowledgments
The authors gratefully acknowledge Dr. Nathaniel Cady for use
of equipment, laboratory space, and reviewing the manuscript. We
thank Dr. Andres Melendez for discussions regarding porphyrin-
based antioxidants. The authors would like to thank Southern Clay
Products for donating samples of Cloisite Na+
. This work was
funded by the new faculty research and FRAP-A awards (University
at Albany, NY).
References
[1] L. Ukrainczyk, M. Chibwe, T.J. Pinnavaia, et al., Environ. Sci. Technol. 29 (2)
(1995) 439–445.
[2] S. Takagi, M. Eguchi, D.A. Tryk, et al., J. Photochem. Photobiol. C: Photochem.
Rev. 7 (2–3) (2006) 104–126.
[3] S. Takagi, T. Shimada, M. Eguchi, et al., Langmuir 18 (6) (2002) 2265–2272.
[4] M. Eguchi, H. Tachibana, S. Takagi, et al., Res. Chem. Intermediat. 33 (1/2)
(2007) 191–200.
[5] G. Magadur, J.-S. Lauret, V. Alain-Rizzo, et al., ChemPhysChem 9 (9) (2008)
1250–1253.
[6] T. Shimidzu, H. Segawa, F. Wu, et al., J. Photochem. Photobiol. A: Chem. 92 (1–
2) (1995) 121–127.
[7] M. Gouterman, The Porphyrins, vol. 3, Academic Press, New York, 1978.
[8] A. Stone, E.B. Fleischer, J. Am. Chem. Soc. 90 (11) (1968) 2735–2748.
[9] Z. Chernia, D. Gill, Langmuir 15 (5) (1999) 1625–1633.
[10] C. Bruckner, P.C.D. Foss, J.O. Sullivan, et al., Phys. Chem. Chem. Phys. 8 (20)
(2006) 2402–2412.
[11] I. Gupta, M. Ravikanth, J. Photochem. Photobiol. A: Chem. 177 (2–3) (2006)
156–163.
[12] S. Takagi, M. Eguchi, D.A. Tryk, et al., Langmuir 22 (4) (2006) 1406–1408.
[13] F. Bedioui, Coordin. Chem. Rev. 144 (1995) 39–68.
[14] S.K. Nobuo Ogata, Takashi Ogihara, J. Appl. Polymer Sci. 66 (3) (1997) 573–581.
[15] L. Granasy, T. Pusztai, J.A. Warren, et al., Nat. Mater. 2 (2) (2003) 92–96.
[16] C. Aguzzi, P. Cerezo, C. Viseras, et al., Appl. Clay Sci. 36 (1–3) (2007) 22–36.
[17] V.G.T. Kuykendall, J. Kerry, Langmuir 6 (8) (1990) 1350–1356.
[18] L. Ukrainczyk, M. Chibwe, T.J. Pinnavaia, et al., J. Phys. Chem. 98 (10) (1994)
2668–2676.
[19] K.A. Carrado, R.E. Winans, Chem. Mater. 2 (3) (1990) 328–335.
[20] M. Ravikanth, D. Reddy, A. Misra, et al., J. Chem. Soc. Dalton Trans. 7 (1993)
1137–1141.
[21] R.S. Czernuszewicz, K.A. Macor, X.Y. Li, et al., J. Am. Chem. Soc. 111 (11) (1989)
3860–3869.
[22] H. van Damme, M. Crespin, F. Obrecht, et al., J. Colloid Interface Sci. 66 (1)
(1978) 43–54.
[23] J-H. Choy, M. Park, Clay Surf. 1 (2004) 403–424.
[24] M. Meot-Ner, A.D. Adler, J. Am. Chem. Soc. 97 (18) (1975) 5107–5111.
[25] P. Moura Dias, D.L.A. de Faria, V.R.L. Constantino, Clays Clay Miner. 53 (2005)
361–371.
[26] A. Ceklovsky´ , A. Czı´merová, M. Pentrák, et al., J. Colloid Interface Sci. 324 (1–2)
(2008) 240–245.
Z. Rice, M. Bergkvist / Journal of Colloid and Interface Science 335 (2009) 189–195 195