close

Вход

Забыли?

вход по аккаунту

?

Peptide surfactants (Pepfactants) for switchable foams and emulsions.

код для вставкиСкачать
ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING
Asia-Pac. J. Chem. Eng. 2007; 2: 362–367
Published online 6 August 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI:10.1002/apj.066
Research Article
Peptide surfactants (Pepfactants) for switchable foams
and emulsions
Andrew S. Malcolm, Annette F. Dexter and Anton P. J. Middelberg*
Centre for Biomolecular Engineering, Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, St Lucia QLD
4072, Australia
Received 1 November 2006; Revised 5 March 2007; Accepted 5 March 2007
ABSTRACT: We have developed a series of sustainable peptide surfactants (Pepfactants) capable of stabilizing foams
and emulsions in a stimuli-responsive manner, based on the reversible formation of a mechanically strong interfacial
film. Under conditions where the interfacially adsorbed peptide forms a mechanically strong ‘film state’, foam or
emulsion stabilization occurs as a direct result of the film strength. Under conditions where the interfacially adsorbed
peptide forms a mobile ‘detergent state’, foam or emulsion stabilization is either reduced, or does not occur. Preformed
foams or emulsions stabilized by Pepfactants undergo rapid phase coalescence when the film state is converted to the
detergent state. Switching between film and detergent states is readily and reversibly achieved by a change in the bulk
solution composition, such as a change in pH, or the addition or sequestering of metal ions.  2007 Curtin University
of Technology and John Wiley & Sons, Ltd.
KEYWORDS: peptide; protein; surfactant; emulsion; foam; Pepfactants; coalescence
INTRODUCTION
Designer peptides have been developed recently as
building blocks for novel self-assembled materials with
stimuli-responsive properties (Fairman and Akerfeldt,
2005). Self-assembly relies on weak noncovalent interactions such as electrostatic interactions, hydrogen
bonding, hydrophobic interactions, metal coordination
and van der Waals interactions. The reversible nature of
these weak interactions facilitates the design of stimuliresponsive materials that reorganize their supramolecular architecture in response to changes in environmental
conditions such as pH, salt or temperature. To date,
such materials have been based on self-assembly in
bulk aqueous solution, for example peptides that gel
in response to pH (Schneider, et al ., 2002) or at solid
interfaces, for example, peptide layers that restructure
in response to pH (Stevens, et al ., 2004).
We recently reported the first stimuli-responsive
material based on a designed peptide at a fluid–fluid
interface (Dexter et al ., 2006). Switching the supramolecular architecture of the adsorbed peptide in
response to changes in bulk solution leads to control of
*Correspondence to: Anton P. J. Middelberg, Centre for Biomolecular Engineering, Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, St Lucia QLD 4072,
Australia. E-mail: cbe@uq.edu.au
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
the mechanical properties of the fluid–fluid interface.
Control of fluid interfaces is of importance in applications using emulsions and foams, where a surface-active
species (surfactant) must be present to inhibit coalescence of the dispersed phases (oil and water, or air
and water). Emulsions and foams are used extensively
across the chemical and process industries including in
the recovery of mineral ores, production of food and
cosmetics, extraction of crude oil, waste water treatment, and pharmaceutical formulation.
AM1 is one member of a novel class of stimuliresponsive peptide surfactants (Pepfactants) that can be
reversibly switched between an interfacial ‘film state’
with significant mechanical strength, and a ‘detergent
state’ in which the molecules do not form a cohesive ensemble. The two states can be actively and
reversibly switched. The film state of the peptide,
but not the detergent state, is effective in stabilizing both foams and oil-in-water emulsions. Emulsions
and foams stabilized by the peptide in the film state
can be rapidly and completely coalesced by changing bulk solution conditions to effect a switch to
the detergent state. This new class of bioproducible
and biodegradable surfactants promises enhanced interfacial control across the widespread applications of
emulsions and foams in the chemical and process
industries.
Asia-Pacific Journal of Chemical Engineering
MATERIALS AND METHODS
General
Reagents were of analytical grade. Ultrapure water
(MilliQ) had a resistivity of 18.2 M cm. Peptide AM1
(amino acid sequence Ac-MKQLADS LHQLARQ
VSRLEHA-CONH2 ) was synthesized by Genscript
Corporation. The purity was >95% by RP-HPLC. The
peptide content was determined by quantitative amino
acid analysis (Australian Proteome Analysis Facility,
Sydney). β-Lactoglobulin (β-LG, 3 × crystallized and
lyophilized) was from Sigma. Sodium dodecyl sulfate
(SDS, biotechnology grade) was from Amresco (Solon,
Ohio, USA). Heavy Middle Eastern sour crude was a
gift from BP Australia.
Interfacial mechanical properties
The Cambridge Interfacial Tensiometer (CIT) (Jones
and Middelberg, 2002) was used to study the mechanical properties of self-assembled AM1 architectures at
the air–water interface. The CIT monitors the transmission of force through an interfacial architecture located
between two anchors floating on a test solution at an
initial separation of 1000 µm. Movement of one anchor
(attached to a piezoelectric motor) away from the other
subjects the interface between the anchors to tensile
strain. Force is registered at the second anchor (attached
to a sensitive force transducer) if a cohesive interfacial
film is present between the anchors. The instrument is
able to determine full interfacial stress–strain curves to
high strain in a cyclic fashion, and does not rely on
assumptions as to the viscous or elastic behavior of the
interface.
To measure interfacial film strength, a solution
(6.5 ml) of surfactant (AM1, SDS or β-LG)
(12.5 µg/ml) in 25 mM sodium 4-(2-hydroxyethyl)-1piperazine ethanesulfonate (HEPES), pH 7.4 or 3.6, was
filled into the CIT bath. For the formation of a cohesive interfacial film, ZnSO4 (100 µM) was included in
AM1-containing buffers. The interface was aged 60 min
before being subjected to tension-compression cycles to
5 and 300% strain. To switch preformed AM1 interfaces, a small volume (≤1% of bath volume) of a stock
of H2 SO4 , NaOH, ZnSO4 or Na+ ethylenediaminetetraacetate (EDTA) was added to the bath by pipetting
beneath the T-pieces. The interface was then allowed to
age 60 min before measurements were repeated.
PEPFACTANTS FOR SWITCHABLE FOAMS AND EMULSIONS
HEPES, pH 7.4 or 3.6. AM1 solutions also contained
200 µM ZnSO4 . Drops (ca. 15 µl) of surfactant solution were formed in air in a humidity chamber. The
drop shape was monitored automatically over 10 min
(DSA-10, Krüss). Control experiments with buffer in
the absence of surfactant showed a stable interfacial
tension close to 73 mN/m.
Foam formation and switching
Surfactant solutions (AM1, SDS or β-LG) were 0.15 or
0.30 mg/ml in 25 mM HEPES pH 7.4. AM1 solutions
also contained 200 µM ZnSO4 . An aliquot (1 ml) of
surfactant was transferred to a custom foam preparation
apparatus, and foam of approximately 6.5 cm height
was formed by passing air (7 ml) upward through the
solution via a sintered glass disk. Foam stability over
2–10 min was monitored photographically. Switching
of AM1-containing foams was achieved by pipetting
an aliquot of H2 SO4 (12.5 µl, 1.6 N) or EDTA (5 µl,
100 mM, pH 8.0) onto the top of the foam. Reverse
switching of AM1 solutions was achieved by adding
an aliquot of NaOH (20 µl, 1 N) or ZnSO4 (30 µl,
23.6 mM).
Emulsion formation and switching
To a solution (2.8 ml) of AM1 (0.15 mg/ml in 25 mM
HEPES, 250 µM ZnSO4 , pH 7.4) was added 0.7 ml
toluene to give an oil fraction of 20% (v/v). To facilitate
visualization of emulsion switching, Sudan III (50 µM)
was included in the oil phase and methylene blue
(10 µM) was included in the aqueous phase. The mixture
was homogenized for 3 min at 24 000 rpm in a rotorstator apparatus (Ystral X10). After homogenization,
aliquots (1 ml) of the emulsions were transferred into
glass vials and stirred magnetically. To switch the
emulsions, an aliquot (8 µl) of acid (1.9 M H2 SO4 ) or
chelating agent (100 mM EDTA, pH 7.6) was added.
Emulsion switching was shown by gross separation of
the oil and water phases. A control vial demonstrated
stability of the emulsion during the test period. In
control experiments, EDTA was included to a final
concentration of 100 µM in the absence of added metal
ions, or H2 SO4 (1.9 M) was added at 1% of the aqueous
phase volume, prior to homogenization.
RESULTS AND DISCUSSION
Interfacial tension measurements
Axisymmetric drop shape analysis employed surfactant (AM1, SDS or ß-LG) at 0.15 mg/ml in 25 mM
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
AM1 is a designed peptide based on the sequence Lac21
(Ac-MKQLADS LMQLARQ VSRLESA-CONH2 )
(Fairman, et al ., 1995), which adsorbs at a fluid–fluid
interface (Middelberg et al ., 2000) to yield a detergent
Asia-Pac. J. Chem. Eng. 2007; 2: 362–367
DOI: 10.1002/apj
363
364
A. S. MALCOLM, A. F. DEXTER AND A. P. J. MIDDELBERG
state incapable of transmitting force in the plane of the
interface (Jones and Middelberg, 2002). AM1 differs
from Lac21 only in the incorporation of two metalbinding histidine (H) residues at positions 9 and 20.
Adsorption of Lac21 or AM1 at an oil–water or
air–water interface is predicted to occur with the
peptide in an α-helical conformation in which repeating
hydrophobic residues at positions 1, 4, 8, 11, 15
and 18 (methionine (M), leucine (L) and valine (V))
form a single hydrophobic face oriented away from
the bulk aqueous phase. This interfacial orientation
has recently been confirmed by neutron reflectivity
studies (Middelberg et al ., 2007). In this orientation, the
histidine residues of AM1 are in contact with the bulk
aqueous phase, but directed toward neighboring peptide
molecules at the interface, enabling the possibility of
metal ion-mediated cross-linking between interfacially
adsorbed peptide molecules.
We used the CIT to characterize the mechanical
properties of the interfacial architecture self-assembled
from AM1 at the air–water interface. At neutral pH
and in the absence of added metal ions, the peptide
ensemble transmitted only minimal force in the plane
of the fluid–fluid interface. The interfacial modulus of
elasticity (initial slope of the interfacial stress vs strain
curve in the low-strain elastic region) was 33 mN/m
and the maximum interfacial stress was 1.4 mN/m
(Fig. 1). The results were similar to those reported for
the parent sequence Lac21 in the absence of added
metal ions (Jones and Middelberg, 2002), and were also
comparable to the behavior of the low molecular weight
surfactant (LMWS) SDS (Fig. 2). When 200 µM Zn(II)
was added to the bulk aqueous solution containing
AM1, a cohesive film state was formed that could
transmit significant force in the plane of the interface
(interfacial modulus of elasticity 93 mN/m, maximum
interfacial stress 7.1 mN/m) (Fig. 1). The mechanical
properties of the AM1 film in the presence of Zn(II)
were comparable to the protein β-LG (Fig. 2).
Reversible switching of the interfacial architecture
was demonstrated by adding sufficient H2 SO4 to shift
the pH of the bulk solution to 3.6, resulting in
the loss of force transmission (interfacial modulus
of elasticity 26 mN/m, maximum interfacial stress of
0.6 mN/m) (Fig. 1). Subsequent neutralization of the
solution returned the interfacial architecture to the film
state (interfacial modulus of elasticity 84 mN/m, maximum interfacial stress 7.1 mN/m) (Fig. 1). Subsequent
addition of the chelating agent EDTA also caused film
dissipation (not shown).
To verify that enhancement of force transmission
on addition of zinc was due to the inclusion of the
histidine residues in the AM1 sequence, we tested the
parent sequence Lac21 in the presence of zinc. Lac21
adsorbed at the interface in the presence of 100 µM
Zn(II) failed to show significant force transmission,
yielding a detergent state with an interfacial modulus of
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pacific Journal of Chemical Engineering
Figure 1. Switching of the mechanical properties of selfassembled AM1 architectures at the air–water interface.
The interface was aged 1 h before each measurement. The
initial solution contained 12.5 µg/ml AM1 at pH 7.4. ZnSO4
was then added to 100 µM, followed in the next step by
H2 SO4 to adjust the pH to 3.6. In the final step, NaOH was
added to restore the pH to 7.4.
Figure 2. Mechanical properties of self-assembled β-LG and
SDS architectures at the air–water interface. The interface
was aged for 1 h before each measurement. Solutions were
at pH 7.4 with surfactant concentrations of 12.5 µg/ml.
elasticity of only 7 mN/m and a maximum interfacial
stress of 0.2 mN/m (data not shown).
The presence of a surfactant to reduce interfacial tension at a freshly formed (bare) interface is necessary
for preparation of thermodynamically unstable dispersions such as foams and emulsions (Weaire and Hutzler,
1999). A strong correlation has been found to exist
between the extent of foam formation and the initial
rate of decrease in interfacial tension (Wilde, 2000).
We tested the interfacial tension behaviour of AM1 at
the air–water interface using pendant drop tensiometry, and contrasted it with the LMWS SDS and the
protein β-LG (Fig. 3). Peptides, like LMWS, are significantly smaller than proteins. AM1 and SDS adsorbed
rapidly at the interface, causing surface tension to drop
Asia-Pac. J. Chem. Eng. 2007; 2: 362–367
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
from that of a clean air–water interface faster than
the shortest time resolution of this technique (∼2 s).
β-LG adsorption was considerably slower, and hence
the interfacial tension was much higher at early times,
in the period relevant to foam formation. AM1 in the
presence of zinc at neutral and acid pH displayed identical surface activity, reaching an interfacial tension of
48 mN/m after 10 min aging. Formation of a detergent
state rather than a film state was thus not accompanied
by significant changes in interfacial tension. Neutron
reflectivity measurement of AM1 interfacial architectures self-assembled in the presence of either Zn(II)
or EDTA indicate that both ensembles consist of an
approximate monolayer of peptide having similar surface excess (Middelberg et al ., 2007). The difference
between the film and detergent states, therefore, appears
to be due to changes in the interaction between peptide
molecules at the interface, effected by the binding of
Zn(II).
The foaming properties of AM1 in the presence of
zinc at neutral and acid pH were compared with SDS
and β-LG. Foams freshly formed from AM1–Zn(II)
solutions at neutral or acid pH (Fig. 4(A), (B)) were
similar to those formed from SDS (Fig. 4(D)) and were
of a higher quality than the β-LG foam (Fig. 4(C)), as
characterized by a smaller average bubble size and a
higher liquid hold-up in the foam column. The denser
foam formation with AM1 and SDS as compared to
β-LG is consistent with the dynamic interfacial tension
results (Fig. 3).
Lowered interfacial tension is not sufficient for foam
stability (Morrison and Ross, 2002). Foams break down
by the parallel processes of drainage, disproportionation and coalescence. Factors that are known to stabilize
foams include interfacial rheological properties such as
surface elasticity and viscosity (Schramm, 2005). It is
believed that a viscoelastic interfacial film diminishes
propagation of surface ripples that lead to film rupture between neighboring foam bubbles (Damodaran,
2005). SDS, despite initially lowering the interfacial
tension much more rapidly and to a greater extent than
β-LG (Fig. 3), was unable to sustain a foam at the concentration tested. Collapse was complete within 2 min
(Fig. 4(D)). At acid pH, where AM1 forms a detergent state equivalent to that observed for SDS, an
AM1–Zn(II) solution was unable to form a long-lasting
foam (Fig. 4(B)). In contrast, the foams formed from
β-LG (Fig. 4(C)) or AM1–Zn(II) (Fig. 4(A)) persisted
after 2 min aging. Both of these surfactant solutions
form a mechanically strong film state at the air–water
interface (Fig. 1).
Increasing the concentration of peptide led to even
greater foam stability. An AM1 foam formed from
a 0.30 mg/ml peptide solution at neutral pH in the
presence of Zn(II) could be aged for 10 min with very
little sign of coarsening or coalescence. However, the
foam collapsed in less than 2 min on addition of a small
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
PEPFACTANTS FOR SWITCHABLE FOAMS AND EMULSIONS
Figure 3. Interfacial tension of surfactant solutions at the
air–water interface. Each surfactant was at 0.15 mg/ml.
AM1–Zn(II) at pH 7.4 (open triangles); AM1–Zn(II) at pH 3.6
(open squares); SDS at pH 7.4 (open circles); β-LG at pH 7.4
(closed triangles).
volume of H2 SO4 or EDTA to the foam column (data
not shown). A new foam formed from the acidified
solution, or solution to which a chelating agent had
been added, collapsed within 2 min, consistent with
self-assembly of AM1 in a mobile detergent state,
rather than a cohesive film state. However, it was
possible to restore foam stability by neutralizing the
bulk solution, or adding Zn(II) in excess of the added
EDTA, respectively, before forming a new foam (data
not shown). The result demonstrates the reversible
nature of peptide interfacial switching.
The results obtained at the air–water interface suggested that switching of AM1 architectures at oil–water
interfaces, such as those present in emulsions, should
also be possible. To test the effects of the stimuliresponsive peptide film on emulsion stability, a 20%
(v/v) toluene-in-water emulsion was prepared using
micromolar AM1 and Zn(II). Under these conditions,
AM1 proved to be an effective emulsifying agent.
The AM1–Zn(II) emulsion was stable to coalescence
over 20 h on standing. However, when excess EDTA
was added to sequester zinc, the emulsion coalesced
on a time scale of seconds (Fig. 5). Similarly, if
an aliquot of H2 SO4 was added to disrupt metalhistidine binding, the peptide emulsion rapidly coalesced, with phase separation commencing in seconds
(not shown). Complete clearing of both phases was
apparent after standing several minutes to allow creaming and coalescence of the remaining oil phase. For
comparison, when an AM1-containing toluene-in-water
emulsion was prepared at neutral pH in the presence of EDTA and the absence of added metal ions,
the dispersion coalesced within seconds after the end
of mixing. Similarly, when an AM1–Zn(II) solution
was acidified before homogenization with toluene, coalescence of the freshly prepared dispersion occurred
within seconds. The emulsion-stabilizing properties of
Asia-Pac. J. Chem. Eng. 2007; 2: 362–367
DOI: 10.1002/apj
365
366
A. S. MALCOLM, A. F. DEXTER AND A. P. J. MIDDELBERG
the peptide thus appear to correlate uniquely with
the interfacial film state of AM1, not the detergent
state.
The suitability of Pepfactants for use in complex
emulsion applications was evaluated by testing the
emulsion-stabilizing and switching behavior of AM1
with a heavy sour crude oil. Preparation of a crude oilin-water emulsion using AM1–Zn(II) as emulsifying
agent gave a low-viscosity emulsion (Fig. 6). In contrast, high-speed mixing of the same oil with water in
the absence of AM1 led to formation of an intractable
sludge. Addition of a small aliquot of acid to the AM1stabilized crude oil emulsion led to phase separation on
a time scale of seconds, as shown in Fig. 6. Similarly,
Asia-Pacific Journal of Chemical Engineering
Breaking of an AM1–Zn(II) toluene-in-water
emulsion by addition of EDTA. The initial pH was 7.4.
No additions were made to the left-hand vial. An aliquot of
EDTA was added to the right-hand vial, while stirring. The
extent of coalescence is shown (a) before additions, (b) at
30 s, and (c) at 60 s after the addition of acid. This figure is
available in colour online at www.apjChemEng.com.
Figure 5.
Figure 6. Phase separation of an AM1-stabilized crude oilin-water emulsion by addition of acid. No additions were
made to the left-hand vial. An aliquot of H2 SO4 was added
to the right-hand vial. The extent of phase separation is
shown (a), before additions, at (b) 10 s and (c) 60 s after the
addition of acid.
addition of the metal chelating agent EDTA to the
AM1–Zn(II) crude oil emulsion led to rapid phase separation (not shown). AM1 is thus capable of acting as
a stimuli-responsive emulsifier even in the presence of
significant levels of competing surfactant, as in complex
crude oil systems.
CONCLUSION
Figure 4. Foaming properties of AM1 compared with β-LG
and SDS. Each panel shows the foam at 0 and 2 min after
formation by bubbling air through 0.15 mg/ml surfactant
solutions. (a) AM1–Zn(II) at pH 7.4; (b) AM1 at pH 7.4;
(c) β-LG at pH 7.4; (d) SDS at pH 7.4. This figure is available
in colour online at www.apjChemEng.com.
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
Biomolecular engineering has provided an exciting new
approach to the preparation of nanostructured materials
having useful functional properties. We have designed
amphipathic peptides that self-assemble to form a
stimuli-responsive interfacial architecture at fluid–fluid
interfaces. Under conditions where the interfacially
adsorbed peptide adopts a cohesive film state, the
peptide surfactants stabilize high-quality emulsions and
foams. However, a change in solution conditions causes
the architecture of the adsorbed peptide to switch to a
mobile detergent state, resulting in rapid breaking of the
emulsion or foam, with phase coalescence occurring on
a time scale of seconds.
Through design on a molecular level to enable control
of mesoscale properties at the interfaces of droplets and
bubbles, we have achieved control over the macroscale
Asia-Pac. J. Chem. Eng. 2007; 2: 362–367
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
properties of bulk emulsions and foams, including those
formed from a complex crude oil.
Pepfactants are produced entirely from a tool kit of
the 20 naturally occurring amino acids, giving a vast
range of potential sequences (20n for a peptide of n
residues in length) that will enable tailoring of Pepfactants for application-specific functionality. The possibility of low-cost bioproduction of peptides (Morreale,
et al ., 2004) together with the inherent biodegradability of biomolecules, offers the prospect of a new and
uniquely sustainable technology utilizing peptides as
designer surfactants.
Acknowledgments
The authors acknowledge financial support from the
Australian Research Council (Grant FF0348465). ASM
acknowledges an Australian Postgraduate Award and
a University of Queensland School of Engineering
Super Scholarship. AFD acknowledges a University of
Queensland Postdoctoral Research Fellowship. APJM
acknowledges an Australian Research Council Federation Fellowship. This research was facilitated by
access to the Australian Proteome Analysis Facility
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
PEPFACTANTS FOR SWITCHABLE FOAMS AND EMULSIONS
established under the Australian Government’s Major
National Research Facilities program. Patent applied
for.
REFERENCES
Damodaran S. J. Food Sci. 2005; 70: R54–R66.
Dexter AF, Malcolm AS, Middelberg APJ. Nat. Mater. 2006; 5:
502–506.
Fairman R, Akerfeldt KS. Curr. Opin. Struct. Biol. 2005; 15:
453–463.
Fairman R, Chao MG, Mueller L, Lavoie TB, Shen LY, Novotny J,
Matsueda GR Protein Sci. 1995; 4: 1457–1469.
Jones DB, Middelberg APJ. Langmuir 2002; 18: 10357–10362.
Middelberg APJ, Radke CJ, Blanch HW. Proc. Natl. Acad. Sci.
U.S.A. 2000; 97: 5054–5059.
Morreale G, Lee EG, Jones DB, Middelberg APJ. Biotechnol.
Bioeng. 2004; 87: 912–923.
Morrison ID, Ross S. Colloidal Dispersions: Suspensions, Emulsions, and Foams. Wiley-Interscience: New York, 2002.
Schneider JP, Pochan DJ, Ozbas B, Rajagopal K, Pakstis L, Kretsinger J. J. Am. Chem. Soc. 2002; 124: 15030–15037.
Schramm LL. Emulsions, Foams and Suspensions; Fundamentals
and Applications. Wiley-VCH: Weinheim, 2005.
Stevens MM, Allen S, Sakata JK, Davies MC, Roberts CJ, Tendler
SJB, Tirrell DA, Williams PM. Langmuir 2004; 20: 7747–7752.
Weaire D, Hutzler S. The Physics of Foams. Oxford University
Press: Oxford, 1999.
Wilde PJ. Curr. Opin. Colloid Interface Sci. 2000; 5: 176–181.
Asia-Pac. J. Chem. Eng. 2007; 2: 362–367
DOI: 10.1002/apj
367
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 216 Кб
Теги
switchable, emulsions, pepfactants, surfactants, foam, peptide
1/--страниц
Пожаловаться на содержимое документа