Document 11559726

advertisement
AN ABSTRACT OF THE DISSERATION OF
Debra K. Gale for the degree of Doctor of Philosophy in Chemical Engineering
presented on July 25, 2011.
Title:
Biomolecule Detection by Amplified Photoluminescence of Germanium
Doped Diatom Biosilica
Abstract approved:
_______________________________________________________________
Gregory L. Rorrer
There is significant interest in the fabrication of nano- and micro-structured
silica with spatially ordered features, which exhibit enhanced optoelectronic
properties with application to the next generation of display devices,
semiconductors, and sensors. Currents methods of fabrication employ top
down processes which use extremes of temperature, pressure, and power.
There is an emergence of interest in bio-fabrication techniques, specifically
diatoms. Diatoms are single celled photosynthetic algae that make silica shells
called frustules. In this study, diatom frustule biosilica acts as a label-free,
photoluminescence based, reporter of immunocomplex formation and is
metabolically doped with germanium (Ge) to amplify the intrinsic baseline
photoluminescence signal. Diatom frustules covalently functionalized with the
Rabbit IgG antibody report immunocomplex formation with the antigen antiRabbit IgG by at least a three times enhancement in the photoluminescence
signal intensity at a surface site density of approximately 4000 IgG molecules
µm-2. Immunocomplex formation on the frustule surface follows a Langmuir
isotherm with a binding constant of 2.8 ± 0.7 x 10-7 M, which is within the
range of binding constants for other conventional detection methods. Diatom
frustule biosilica metabolically doped with up to 0.4 weight % Ge by a two
stage photobioreactor cultivation strategy exhibits amplified
photoluminescence upon annealing in air up to 400°C. X-ray photoelectron
spectroscopy shows that germanium dioxide (GeO2) metabolically doped in the
frustule biosilica is thermally converted to germanium oxide (GeO) which
exhibits an amplified photoluminescent signal by up to four times.
Additionally, Raman spectroscopy mapping demonstrates that the Ge can be
metabolically targeted to a specific sub-micron location in the diatom frustule.
Ge doped diatom biosilica reports biomolecule detection by a more intense
photoluminescence signal intensity by a factor of 10 over the native biosilica
without Ge. This study demonstrates that Ge doped silica structures with
nano- and micro-spatially ordered features can be biologically fabricated to
exhibit enhanced photoluminescence properties for future sensor and display
applications.
© Copyright by Debra K. Gale
July 25, 2011
All Rights Reserved
Biomolecule Detection by Amplified Photoluminescence of Germanium
Doped Diatom Biosilica
by
Debra K. Gale
A DISSERTATION
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Doctor of Philosophy
Presented July 25, 2011
Commencement June 2012
Doctor of Philosophy dissertation of Debra K. Gale presented on July 25,
2011.
APPROVED:
_______________________________________________________________
Major Professor, representing Chemical Engineering
_______________________________________________________________
Head of the School of Chemical, Biological, and Environmental Engineering
_______________________________________________________________
Dean of the Graduate School
I understand that my dissertation will become part of the permanent collection
of Oregon State University libraries. My signature below authorizes release of
my dissertation to any reader upon request.
_______________________________________________________________
Debra K. Gale, Author
ACKNOWLEDGEMENTS
I express sincere appreciation and gratitude to Dr. Gregory L. Rorrer,
my PhD advisor. Dr. Rorrer guided me through the research process and
taught me how to think critically about research problems. Dr. Rorrer gave me
exposure to research centers and instruments at Portland State University,
CAMCOR at the University of Oregon, and EMSL at PNNL. He encouraged
me to network with a broader scientific community by attending 12
conferences and doing an internship at Life Technologies. I really appreciate
Dr. Rorrer’s encouragement and support as a mentor.
I graciously acknowledge the National Science Foundation (Award #
0400684) for research support.
I acknowledge my parents, John and Pam Gale and my husband, Justin
Smith for their support. Additionally, Justin took photographs that went into 2
publications and helped me put together the cover for the Journal of Materials
Chemistry, issue 29.
CONTRIBUTION OF AUTHORS
Timothy Gutu and his advisor Dr. Jun Jiao in the Department of
Physics at Portland State University conducted the electron microscopy work.
Clayton Jeffryes generated diatom biosilica samples for the Journal of
Materials Chemistry study.
Dr. Chih-Hung Chang provided insight into the theory and
experimental measurements of the diatom biosilica optoelectronic properties.
TABLE OF CONTENTS
Page
Chapter 1: Introduction………………………..................................................1
Chapter 2: Photoluminescence Detection of Biomolecules by AntibodyFunctionalized Diatom Biosilica….......................................................17
Abstract.................................................................................................18
Introduction...........................................................................................19
Results...................................................................................................20
Discussion.............................................................................................25
Experimental.........................................................................................29
Chapter 3: Thermal annealing activates amplified photoluminescence of
germanium metabolically doped in diatom
biosilica.................................................................................................49
Abstract.................................................................................................50
Introduction...........................................................................................51
Experimental.........................................................................................52
Results...................................................................................................54
Discussion.............................................................................................59
Chapter 4: Germanium Enrichment in the Girdle Band of the Centric Diatom
Cyclotella sp..........................................................................................77
Abstract.................................................................................................78
Introduction...........................................................................................79
Experimental.........................................................................................80
Results...................................................................................................87
Discussion.............................................................................................93
TABLE OF CONTENTS (Continued)
Page
Chapter 5: Immunocomplex Detection by Photoluminescent Germanium
Centers Doped in Diatom Biosilica....................................................113
Abstract...............................................................................................114
Introduction.........................................................................................115
Experimental.......................................................................................116
Results.................................................................................................119
Discussion...........................................................................................123
Chapter 6: Conclusion....................................................................................138
Bibliography....................................................................................................143
Appendix A: Procedures.................................................................................156
LIST OF FIGURES
Figure
1-1.
1-2.
2-1.
2-2.
2-3.
2-4.
2-5.
2-6.
2-7.
Page
Scanning electron microscope images of diatom frustules. (a)
Cyclotella sp. (b) Coscinodiscus wailessi (c) Pinnularia sp...............15
Schematic of the diatom cell division cycle and frustule formation.....16
Electron microscopy images of a representative biosilica frustule valve
isolated from the centric diatom Cyclotella sp. (a) SEM image of whole
valve, revealing surface topology; (b) TEM image of whole valve,
revealing radial symmetry of the nanopore arrays; (c) TEM image of
nanopore pattern detail..........................................................................40
The basic structure of the antibody-functionalized diatom biosilica (not
drawn to scale)......................................................................................41
FT-IR spectra of frustule biosilica before and after amine
functionalization....................................................................................42
Imaging of functionalized diatom biosilica. (a) Epifluorescent image of
amine-functionalized diatom frustule labeled with fluorescamine; (b)
Epifluorescent image of Rabbit IgG antibody-functionalized diatom
frustule labeled with DyLightTM 488 Goat anti-Rabbit IgG.; (c) TEM
image of amine-functionalized diatom frustule challenged with Goat
anti-Rabbit IgG labeled 10 nm gold nanoparticles (4.5 antigens/gold
nanoparticle); (d) TEM image of antibody-functionalized diatom
frustule challenged with Goat anti-Rabbit IgG labeled 10 nm gold
nanoparticles (4.5 antigens / goldnanoparticle)....................................43
Comparison of surface site density of Goat anti-Rabbit IgG on the
valve of bare diatom biosilica, amine-functionalized diatom biosilica
(diatom NH2-biosilica), and Rabbit IgG antibody functionalized diatom
biosilica (diatom+antibody)..................................................................44
Comparison of photoluminescence (PL) spectra of bare diatom
biosilica, Rabbit IgG antibody-functionalized diatom biosilica
(diatom+antibody), and Rabbit IgG immunocomplex functionalized
diatom biosilica (diatom+antibody+antigen). Excitation wavelength
was 337 nm. Inset: photograph of PL emission from antibodyfunctionalized diatom biosilica.............................................................45
Comparison of normalized peak PL intensity of antibody functionalized
diatom biosilica with that of an antibody-functionalized, thermally
grown SiO2 thin film. (a) Bare diatom biosilica, Rabbit IgG antibodyfunctionalized diatom biosilica (diatom+antibody), Rabbit IgGfunctionalized diatom biosilica with Goat anti-Rabbit IgG
(diatom+antibody+antigen), and Rabbit IgG antibody-functionalized
LIST OF FIGURES (Continued)
Figure
2-8.
2-9.
3-1.
3-2.
3-3.
3-4.
3-5.
3-6.
Page
diatom biosilica challenged with a non-complimentary Goat antiHuman IgG (diatom+antibody+non-comp. antigen). (b) Thermally
grown SiO2, Rabbit IgG antibody-functionalized SiO2 (SiO2
+antibody), and Rabbit IgG immunocomplex functionalized SiO2
(SiO2+antibody+antigen). Excitation wavelength was 337 nm............46
Peak PL intensity of Rabbit IgG antibody functionalized biosilica vs.
Goat anti-Rabbit IgG antigen concentration. Solid line presents best fit
to Equation (1)......................................................................................47
This work was featured on the inside cover of Advanced Functional
Materials, volume 19, number 6, 2009.................................................48
(a) Cell number density and soluble Si concentration in culture medium
as a function of time for the two-stage photobioreactor cultivation of
Pinnularia sp. (b) Soluble Si and Ge concentration in culture medium
of Stage II for an initial Ge concentration of 25 µm.............................69
TEM images of Pinnularia sp. obtained at the end of stage II
cultivation where no Ge was fed to the culture, the cells were silicon
starved, and the cell number density was constant for two photoperiods.
(a) Intact frustule; (b) submicron pore structure; (c) nanoscale pore
array. ....................................................................................................70
TEM images and electron diffraction of Pinnularia sp. obtained at the
end of the stage II cultivation which was doped with 0.49 weight % Ge
and annealed in air at 400ºC for 2 hours. (a),(b) Microstructure of
Pinnularia sp. frustule; (c),(d) Patterned submicron pore array; (e),(f)
and nanoscale pore features with electron diffraction pattern
inset.......................................................................................................71
Photoluminescence and thermal gravimetric analysis (TGA) as a
function of annealing temperature of (a) biosilica doped with 0.45 wt %
Ge and (b) native biosilica without Ge. (c) TGA profile of native
biosilica weight loss percent as a function of time for biosilica without
Ge. Photoluminescence signal normalized to the signal prior to
annealing...............................................................................................72
(a) Normalized photoluminescence intensity and (b) peak
photoluminescence intensity wavelength of biosilica with 0, 0.24, 0.46,
0.49 and 0.96 wt % Ge in silica as a function of thermal annealing in air
for 2 hours at 250ºC, 400ºC, and 600ºC................................................73
Normalized photoluminescence spectrum of native biosilica without Ge
and biosilica doped with 0.96 weight percent Ge, which was annealed
LIST OF FIGURES (Continued)
Figure
3-7.
3-8.
4-1.
4-2.
4-3.
4-4.
4-5.
4-6.
4-7.
Page
at 250ºC. The photoluminescence spectrum of the native biosilica was
multiplied by a factor of 5 for scale. Inset shown is a photograph of
Pinnularia frustule powder annealed at 250°C excited with 337 nm
light.......................................................................................................74
(a) X-ray photoelectron spectroscopy (XPS) analysis of industrial GeO2
and Pinnularia sp. biosilica doped with 0.96 wt % Ge which was as
deposited (no annealing), and annealed in air for 2 hours at 250ºC,
400ºC, and 600ºC. (b) XPS peak height intensity ratio of GeO to GeO2
before and after annealing. The binding energy of GeO2 is 1220.2 eV
and the binding energy of GeO is 1221.6 eV........................................75
This work was features on the front cover of the Journal of Materials
Chemistry, volume 21, number 29, 2011..............................................76
a) Cell number density and soluble Si concentration in the culture
medium of Stage II for a 48 hour addition of Si and Ge. (b) Cell
number density and soluble Si concentration in the culture medium of
Stage II for a 48 hour addition of Si....................................................102
Si and Ge concentration profile in photobioreactor medium during
Stage II and concentration profile in the culture medium if there was no
uptake. (a) Profile for Si and Ge addition in Stage II which corresponds
to Figure 1a. (b) Profile for Si addition in Stage II which corresponds to
Figure 1b.............................................................................................103
(a) A schematic of the diatom cell cycle which correlates to the cell
density and Ge uptake as a function of cultivation time. (b) Predicted
Ge/Si mass ratio in the frustule valve and girdle band as a function of
cultivation time...................................................................................104
STEM-EDS spot scan analysis of an intact biosilica frustule valve for
the detection of Ge. (a) Spot denoted with a 1 indicates the location of
analysis. (b) EDS energy spectrum which confirmed the metabolic
insertion of Ge with the Kα energy peak at 9.86
keV......................................................................................................105
TEM images of an (a) intact biosilica frustule and the (b) submicron
patterned pore structure of the frustule valve obtained at the end of
Stage I prior to Ge and Si addition in Stage II....................................106
TEM images of an (a) intact biosilica frustule,(b),(c) submicron
patterned pore structure, and (d) porous nanostructure of the frustule
valve obtained at 120 hours into Stage II addition of Si and Ge........107
(a) SEM image of a Cyclotella frustule which shows the parent valve
(epivalve), parent girdle band (epicingulum), daughter valve
LIST OF FIGURES (Continued)
Figure
Page
(hypovalve), and daughter girdle band (hypocingulum). (b) Schematic
of a diatom frustule which shows the dimensions used for model
development. (c) Schematic of the diatom frustule hypotheca at the
end of Stages IIA and IIB which shows Ge enrichment in the daughter
girdle bands.........................................................................................108
4-8. (a) Raman spectrum of frustule biosilica doped with Ge. (b) Raman
spectrum of Ge doped biosilica showing SiO2 peak (490 cm-1) and
GeO2 peak (420 cm-1). (c) Raman spectrum of biosilica without
Ge........................................................................................................109
4-9. Representative Raman line scans of normalized Raman intensity of
GeO2 (420 cm-1) to SiO2 (970 cm-1). (a), (b) Ge doped biosilica. (c)
Frustule valve and girdle band from Stage II fed Ge, which did not
have Ge. (d) Frustule from Stage II which was only fed Si. (c-inset)
Representative biosilica frustule for Raman spectroscopy
analysis................................................................................................110
4-10. Raman map of intact Cyclotella sp. frustule. (a) Blue and (b) red set to
the silica raman bands of 970 cm-1 and 490 cm-1. (c) GeO2 Raman band
of 420 cm-1 set to green.......................................................................111
5-1. a) Cell number density and soluble Si concentration in the culture
medium of Stages I and II for a 48 hour addition of Si and Ge starting
at 131 hours indicated by the dotted line. (b) Si and Ge concentration
profile in the culture medium during Stage II and Si and Ge
concentration profile in the culture medium if there was no uptake. .132
5-2. TEM images of the valve facing view of an intact diatom frustule and
the submicron patterned pore detail on the frustule valve. Frustules
were extracted from Cyclotella diatom cells (a), (b) at the end of Stage
I prior to Si and Ge addition and (b), (c) 120 hrs into Stage II...........133
5-3. Normalized photoluminescence intensity of biosilica from Stage II of
the cultivation which was fed Ge and the Ge weight percent in the
biosilica. Biosilica was annealed in air at 400°C for 2 hours, then
functionalized with an antibody. The photoluminescence intensity is
normalized to the photoluminescence intensity of each sample prior to
annealing and functionalization treatment..........................................134
5-4. Epifluorescence image of a diatom frustule acquired 120 hours into Stage
II of Si and Ge addition which was annealed at 400°C in air for 2 hours
then functionalized with a Rabbit IgG antibody and a fluorescein labeled
goat anti-Rabbit IgG antigen.................................................................135
LIST OF FIGURES (Continued)
Figure
Page
5-5. Normalized photoluminescence intensity of biosilica with 0 and 0.4 wt.
% Ge which was functionalized with an antibody and immunocomplex
with 28.6 μM (5000x dilution) or 143 mM (1x dilution) APS
concentration.......................................................................................136
5-6. a) Normalized photoluminescence intensity of annealed biosilica with
0 and 0.4 wt. % Ge which was functionalized with an immunocomplex
using a 28.6 μM APS concentration...................................................137
A-1. Photograph of the photoluminescence system in Gleeson laboratory
308.......................................................................................................197
A-2. Pinnularia diatom deposition on PDMS coated glass slide................199
LIST OF TABLES
Table
3-1.
Page
Process parameters for two-stage photobioreactor cultivation of
Cyclotella sp........................................................................................112
LIST OF APPENDIX PROCEDURES
Procedure
Page
A-1. Amine Functionalization of H2O2 Treated Diatom Frustules
with 3-aminopropyltrimethoxysilane (1X APS Dilution,
0.143 mol APS/L) .............................................................................157
A-2. Amine Functionalization of H2O2 Treated Diatom Frustules
with 3-aminopropyltrimethoxysilane (5000x APS Dilution,
2.864x10-5 mol APS/L) .....................................................................159
A-3. Antibody Rabbit Immunoglobulin G (IgG) Crosslinking of
Bis[sulfosuccinimidyl]suberate (BS3),
3-aminopropyltrimethoxysilane (APS) Functionalized
Diatom Frustule Films on Round Coverslips .....................................161
A-4. Antibody Rabbit Immunoglobulin G (IgG) Crosslinking of
Bis[sulfosuccinimidyl]suberate (BS3), 3-aminopropyltrimethoxysilane
(APS) Functionalized Diatom Frustules.............................................163
A-5. Antigen Rabbit Anti-IgG Crosslinking of 3aminopropyltrimethoxysilane (APS), Bis[sulfosuccinimidyl]suberate
(BS3), Rabbit IgG Functionalized Diatom Frustule Films on Round
Coverslips............................................................................................164
A-6. Antigen Rabbit Anti-IgG Crosslinking of 3aminopropyltrimethoxysilane (APS), Bis[sulfosuccinimidyl]suberate
(BS3), Rabbit IgG Functionalized Diatom Frustules..........................167
A-7. Harrison’s and Guillard’s f/2 enrichment Artificial Seawater
Medium...............................................................................................170
A-8. Avidin-Fluorescein (FITC) Functionalization of Biotinylated Diatom
Frustules..............................................................................................177
A-9. Biotin Functionalization of Amine Functionalized Diatom
Frustules..............................................................................................178
A-10. Bis[sulfosuccinimidyl]suberate (BS3) Crosslinking of 3aminopropytrimethoxysilane (APS) Functionalized Diatom
Frustules..............................................................................................179
A-11. Bis[sulfosuccinimidyl]suberate (BS3) Crosslinking of 3aminopropytrimethoxysilane (APS) Functionalized Diatom Frustule
Films on Round Coverslips.................................................................181
A-12. Coscinodiscus wailesii -Rh123 Sample Preparation for Deposition on
PAH Polymer......................................................................................182
A-13. Deposition of Polyectrolyte Multilayer on Microscope Slide............185
APS Functionalized Diatom Frustule Film Preparation for PL
Sensitivity Study.................................................................................187
LIST OF APPENDIX PROCEDURES (Continued)
Procedure
Page
A-14. APS Functionalized Diatom Frustule Film Preparation for PL
Sensitivity Study.................................................................................187
A-15. Fluorescamine Conjugation of Amine Functionalized Diatom Frustule
Films on Round Coverslips.................................................................189
A-16. Fluorescamine Conjugation of Amine Functionalized Diatom
Frustules..............................................................................................191
A-17. Micro Raman Spectroscopy Analysis of Diatom Biosilica................192
A-18. Phosphate Buffered Saline Preparation, 0.01 M Phosphate, 0.15 M
NaCl, pH 7.6.......................................................................................194
A-19. Photoluminescence (PL) System Alignment......................................195
A-20. Pinnularia Diatom Deposition on Poly(dimethylsiloxane) (PDMS)
Coated Glass Slide..............................................................................198
A-21. Pinnularia Diatom Film Deposition on Glass Coverslips for Annealing
and Photoluminescence Measurements..............................................200
A-22. Preparation of Wetted Cyclotella sp. Biosilica for Photoluminescence
Spectroscopy.......................................................................................201
A-23. Preparation of Rabbit Immunoglobulin G (IgG) for Photoluminescence
Spectroscopy.......................................................................................202
A-24. Subculture of Cyclotella Sp................................................................203
A-25. Thermal Gravimetric Analysis (TGA) of Diatom Biosilica...............205
A-26. Thermally Grown SiO2 Preparation for PL Sensitivity Study............207
Biomolecule Detection by Amplified Photoluminescence of Germanium
Doped Diatom Biosilica
Chapter 1: Introduction
Research Rationale
There is significant interest in using silicon based nano- and microparticles, films and meso-structures as transducer platforms for biosensors. [13] Silica can be readily functionalized and due to its surface chemistry, it
biologically benign, non-toxic, water-stable, and can have a large surface area.
Sensor devices which employ a silica platform have used capacitance, [4,5]
resistance, [6,7] optical reflectivity and photoluminescence (PL) [8-10] as the
detection transducer. PL based detection is a promising detection method
because it inherently exhibits a low signal background. [10] However, not all
morphologies of silica are photoluminescent. Silica nano-and micro-particles
must be doped with a luminophore upon synthesis or porous silicon must be
ultrasonicated into smaller, typically non-uniform particles to yield
photoluminescent particles. [2, 11-13] Films are made photoluminescent by
etching crystalline silicon wafers in hydrofluoric acid and depending on the
reaction conditions, the morphology and PL can vary greatly. [14]
Mesoporous silica is intrinsically photoluminescent, however the synthesis
conditions significantly affect the PL, [15] and often it is doped with dyes, tin,
and other luminescent materials to increase the PL intensity. [16,17] To
address the limitations of conventional optoelectronic material fabrication,
there is an emergence of interest in bio-fabbrication techniques to assemble
nano- to micro-scale hierarchical semiconductor materials. [18,19]
Specifically, diatoms have been touted as a paradigm for biological fabrication
of nanostructured silica which is instrinsically photoluminescent. [20,21]
2
Biosensors for Immunocomplex Detection
Biosensors, which measure antibody-antigen immunocomplex
formation, are essential to diagnose and profile diseases rapidly and reliably.
[22] Recently, biosensors based upon semiconductor materials such as silicon
and germanium, have gained considerable attention due to their optoelectronic
and chemical properties. [23-26] These materials show potential as label-free
biological sensors despite their top down fabrication which includes
photolithography, thin film deposition, and etching of a crystalline silicon
wafer. [27] Transduction mechanisms of silicon based sensors include
refractive index, photoluminescence, reflectance, and electrochemical signals.
A limitation to silicon sensor technology is poor sensitivity which makes
miniaturization into an array for multi-immunocomplex detection nearly
impossible. [28] On a large scale, these materials are difficult to fabricate and
integrate into a lab on a chip biosensor device.
The immunocomplex is typically detected and quantified by optical,
electrical, or mass measurement methods using surface-immobilized
biomolecules. [29-31] Electrical and mass measurement methods are often
destructive, invasive, and yield low signal intensities. Furthermore, the labels
used in many optical methods can introduce contamination to the sample
matrix. Biosensing platforms that use label-free optical detection, such as
ellipsometry and surface plasmon resonanace, avoid these limitations.
However, ellipsometry can only sense biomolecules on a flat surface of small
area, and it is difficult to detect multiple biomolecules with surface plasmon
resonance. [31]
In effort to fabricate semiconductor materials with tunable nanoscale
properties, supramolecular patterned three dimensional structures, often
inspired by nature, are being synthesized using peptide mediated bottom up
self-assembly. [32, 33, 18] Materials yet to be studied for use as a biosensor
3
are diatoms, a unique class of algae. Diatoms fabricate from the bottom up,
optoelectronic, three dimensional, silica structures ordered at the nano and
microscales and are a paradigm for controlled fabrication of nanostructured
silica. [34,55]
Below, the biochemistry of biosilica formation in diatoms is described
in order to provide a basis for imparting photoluminescent properties to the
material.
Fabrication, Properties, and Functionalization of Diatom Biosilica
Diatoms are single-celled photosynthetic algae of the class
Bacillariophyceae, which take up soluble silicon in the form of Si(OH)4, silicic
acid from the surroundings and deposit it into a three dimensional shell or
“frustule” which is ordered at the nano and microscales. Biosilica diatom
frustules are made up of two halves, the epivalve and hypovalve, which fit
together like a petri dish and are held together by a girdle band. [36] Diatoms
are classified into two groups: centric, which are radially symmetrical, or
pennate, that are characterized with bilateral symmetry. [36] It is estimated
that over 10,000 different species exist, which range in sizes from 1 μm to 5
mm of varying height, length, diameter, pore size, and pore number. [36-38]
SEM micrographs of the centric diatoms Cyclotella sp. and Coscinodiscus
wailessi in Figure 1. Demonstrate the diversity of diatom biosilica.
Understanding the fabrication of these unique silica frustules by the diatom cell
offers a new way to engineer diatom biosilica structures with interesting
optoelectronic properties.
Due to the obligate silicon requirement by diatoms for cell division and
frustule formation, the cell cycle and silicon deposition are closely tied. [39,40]
Diatom cell division commences with the gap phase (G1) of cell division upon
intake of soluble silicon as shown in Figure 2. [41] During this phase, the
4
cytoplasm volume is increased and organelles double. The girdle bands of the
cyclotella frustule are fabricated. The cell then enters the S phase where the
DNA is replicated and the cytoplasm, which is enclosed by a membrane called
the plasmalemma, separates the cytoplasm into two distinct pockets where it
will continue to increase in volume in the G2 phase. The M phase commences
with the formation of the silicon deposition vesicle (SDV). The SDV is an
acidic compartment formed at the center of cell division in each cytoplasmic
pocket, which spans the width of the cell and is the location for
biosilicification of soluble silicon into the new daughter frustules. [39, 41] The
acidic environment of the SDV contains proteins called silaffins that promote
the condensation of soluble silicon into silica nanoparticles less than 50 nm in
diameter, which ultimately fuse together forming the frustule daughter valve.
[37, 40, 42] The girdle bands are fabricated after the hypovalve in a separate
SDV and upon completion the hypovalve and girdle bands are excised from
the SDV and the daughter cells separate, completing cell division. [41,43]
Cell division and ultimately nanoscale biosilica formation can be
controlled by external manipulation of silicon. Three modes of soluble silicon
uptake occur in diatoms: surge uptake, internally controlled uptake, and
externally controlled uptake. [44-46] Surge uptake occurs when a silicon
starved cell is exposed to silicon. Maximum silicon uptake rates have been
observed in this mode of uptake, and immediately following surge uptake,
frustule deposition begins and pools of silicon are made for storage. During
internally controlled uptake, the concentration of silicon in the cell is high,
such that frustule deposition is limited only by the rate of silicic acid
condensation and polymerization into the frustule. In externally controlled
uptake, the intercellular and extracellular silicon concentrations are low, such
that the rate of biosilica deposition is controlled by the extracellular silicon
5
concentrations. Maintaining the diatom cell in a state close to silicon
starvation imparts control over frustule deposition.
Since the surface of diatom frustules are covered by reactive silanol
groups, functionalization of biosilica with biological molecules with various
reaction chemistries has been shown and validated with fluorescent markers.
[47-49] Additionally, the modulation of the intrinsic diatom biosilica
photoluminescence has been observed upon adsorption of gaseous species such
as nitrogen dioxide, carbon monoxide, and methane. [50-52]
Diatom frustule biosilica, [50, 52-54] like mesoporous silica,[55] silica
nanoparticles [56] and silicon oxide thin films [57] exhibit photoluminescence
when excited in the ultra-violet range, which is attributed to surface defects
including silanol groups, ≡Si-OH, surface terminated hydrogens, ≡Si-H, [58]
and non-bridging oxygen hole centers, ≡Si-O·. [59]
Amplified Photoluminescence of Germanium Doped Diatom Biosilica
Nanostructured silica has received significant attention due to its
visible photoluminescence. [55, 56, 57, 60, 61] Recently there has been
significant interest in doping silica with germanium (Ge), which has a higher
carrier mobility to favricate semiconductor materials with luminescent
properties. [62-65] These materials are of interest because the absorption
properties in the ultra-violet range of Ge doped silica is 2-3 times greater than
that of undoped silica, rendering highly luminescent semiconductor materials.
[66] This luminescent behavior is attributed to germanium oxygen deficient
centers (=Ge which exhibit blue photoluminescence at approximately 410
nm. [62, 67] Ion implantation, magnetron sputtering, sol-gel deposition, and
chemical vapor deposition, followed by annealing, are a sample of the
fabrication techniques used to embed low levels of germanium into silica.[62,
67, 75]
6
Germanic acid, Ge(OH)4 a chemical analog to silicic acid, Si(OH)4 has
been shown to be taken up by diatoms and incorporated into the biosilica
frustule. [68-73] Work in our laboratory showed that by using a two stage
bioreactor cultivation strategy exploiting surge uptake of Si and Ge, the diatom
Pinnularia sp. could incorporate levels of Ge up to 1 weight percent into its
frustule. [72] Additionally, Ge incorporation into the frustule at low levels has
shown to alter the pore structure of the frustule at the micro and nano scales.
[72, 73]
Manipulation of the cell cycle to control Ge uptake into the diatom cell
by externally controlled uptake has not been explored. To harness this type of
substrate uptake by the cell, could be a way to control the time scale of the cell
cycle, which ultimately controls the deposition of the valve and girdle band of
the frustule. This method of Ge incorporation into diatom frustules could be
significant for the enhancement of the frustule photoluminescent properties.
Furthermore, exploitation of the photoluminescence properties of Ge
doped diatom biosilica for immunosensing applications has not been studied.
Work also done in our lab used germanium doped biosilica to fabricate an
electroluminescent device which exhibited superior performance to the device
using biosilica without Ge. [74] This study demonstrated the uniquely
enabling properties of germanium doped biosilica for electronic applications.
However, there has been no research on characterization of the
photoluminescent properties of Ge doped biosilica and how these properties
could be optimized for immunosensing applications.
To date, there have been no studies that harness and tune the intrinsic
photoluminescence of diatom biosilica as a non-labeling transduction
mechanism to measure immunocomplex binding on an antibody functionalized
diatom frustule. Furthermore, each diatom frustule could be considered as one
microscale biosensor that could be incorporated into a lab on a chip device to
7
measure multi-immunocomplex solutions. The intrinsic chemical and
photoluminescent properties of diatom biosilica offer opportunities for the next
generation of sensors and optoelectronic devices.
Research Hypothesis
The photoluminescence of nanostructured diatom biosilica can be
harnessed as a transducer to report immunocomplex formation by enhanced
photoluminescence. Furthermore, doping the diatom biosilica with germanium
to enhance the intrinsic baseline photoluminescence intensity will allow for
enhanced detection.
Research Goals and Objectives
This research will focus on harnessing the instrinsic photoluminescence
of diatom biosilica for biosensing. Towards this goal, I will characterize the
photoluminescence based detection of immunocomplex formation on the
diatom biosilica surface. I will then characterize immunocomplex detection by
germanium doped diatom biosilica. This research has four objectives:
1. Functionalize diatom biosilica with an antibody and detect immunocomplex
formation by enhanced photoluminescence (Chapter 2).
2. Develop a cultivation strategy to dope germanium into the girdle band of
the diatom biosilica frustule (Chapter 4).
3. Activate the photoluminescence of germanium doped in the diatom frustule
(Chapter 3).
4. Characterize immunocomplex detection by enhanced photoluminescence of
diatom biosilica doped with germanium (Chapter 5).
8
This research will utilize the intrinsic photoluminescence properties of
diatom biosilica for immunocomplex sensing. A biosensor device fabricated
with a diatom biosilica platform will enable for label free, stand-off detection
of biological analytes.
Chapter Summary
Chapter 2. Photoluminescence Detection of Biomolecules by AntibodyFunctionalized Diatom Biosilica
In this study, antibody-functionalized diatom biosilica frustules serve as
a microscale biosensor platform for selective and label-free photoluminescence
based detection of immunocomplex formation. The model antibody rabbit
immunoglobulin G (IgG) is covalently attached to the frustule biosilica.
Antibody-antigen (immunocomplex) formation on the diatom biosilica surface
is reported by a 3 times enhancement in the photoluminescence intensity. This
unique photoluminescence enhancement is correlated to the antigen (goat antiRabbit IgG) concentration, where immunocomplex binding follows a
Langmuir adsorption isotherm.
Chapter 3. Thermal Annealing Activates Amplified Photoluminescence of
Germanium Metabolically Doped in Diatom Biosilica
In this study, the photoluminescence of germanium metabolically
doped into diatom biosilica is activated by thermal annealing in air. Activated
photoluminescent germanium in the biosilica results in up to a four times
enhancement in the baseline photoluminescence intensity. Thermal annealing
converts amorphous germanium dioxide metabolically inserted into the
frustule to intensely photoluminescence germanium oxide.
Chapter 4. Germanium Enrichment in the Girdle Band of the Centric
Diatom Cyclotella sp.
9
In this study, a two stage photobioreactor cultivation strategy was
developed to metabolically insert germanium into the girdle band of the centric
diatom frustule cyclotella sp. In stage I, the cells were grown to stationary
phase and silicon starvation. In stage II, the cells were fed a mixture of silicon
and germanium for 48 hours, which was 2 photoperiods. A mathematical
prediction estimated that germanium was enriched in the girdle band by a
factor of three compared to the germanium in the valve. Micro-Raman
mapping confirmed that germanium was enriched in the girdle band of the
diatom frustule.
Chapter 5. Immunocomplex Detection by Photoluminescent Germanium
Centers Doped in Diatom Biosilica
In this study, diatom biosilica doped with photoluminescent germanium
centers, was tested for the ability to report immunocomplex formation by
enhanced photoluminescence. The germanium doped, annealed biosilica
reported antibody functionalization on the surface by a 50 times enhancement
in the PL, although it was unable to report immunocomplex formation. The
antibody concentration was reduced, which tunes the PL signal to report
immunocomplex formation on the diatom biosilica surface. The germanium
doped biosilica reported immunocomplex formation with a photoluminescence
signal which was three times more intense than the PL signal of the native
biosilica without germanium upon immunocomplex formation.
References
[1]
Wang, L.; Zhao, W.; Tan, W. Nano Res. 2008, 1, 99-115.
[2]
Knopp, D.; Tang, D.; Niessner, R. Anal. Chim. Acta 2009, 647, 14-30.
[3]
Gu, Z.; Chen, X.-Y.; Shen, Q.-D.; Ge, H.-X.; Xu, H.-H. Polymer 2010,
51, 902-907.
10
[4]
Schechter, I.; Ben-Chorin, M.; Kux, A. Anal. Chem. 1995, 67, 37273732.
[5]
Zhang, Q.; Shin, Y.J.; Hua, F.; Saraf, L.V.; Matson, D.W. J. Nanosci.
Nantechnol. 2008, 8, 3008-3012.
[6]
Yao, Z.; Yang, M. Sens. Actuators, B 2006, 117, 93-98.
[7]
Lewis, S.E.; DeBoer, J.R.; Gole, J.L.; Hesketh, P.J. Sens. Actuators, B
2005, 110, 54-56.
[8]
Gulino, A.; Giuffrida, S.; Mineo, P.; Purrazzo, M.; Scamporrino, E.;
Ventimiglia, G.; van der Boom, M.E.; Fragalá, I. J. Phys. Chem. B
2006, 110, 16781-16786.
[9]
Ko, M.C.; Meyer, G.J. Chem. Mater. 1996, 8, 2686-2692.
[10]
Sailor, M.J.; Wu, E.C. Adv. Funct. Mater. 2009, 19, 3195-3208.
[11]
Santra, S.; Zhang, P.; Wang, K.; Tapec, R.; Tan, W. Anal. Chem. 2001,
73, 4988-4993.
[12]
Heinrich, J.L.; Curtis, C.L.; Credo, G.M.; Sailor, M.J.; Kavanagh, K.L.
Science 1992, 255, 66-68.
[13]
Bley, R.A.; Kauzlarich, S.M.; Davis, J.E.; Lee, H.W.H. Chem. Mater.
1996, 8, 1881-1888.
[14]
Zhang, X.G. J. Electrochem. Soc. 2004, 151, C69-C80.
[15]
Anedda, A.; Carbonaro, C.M.; Clemente, F.; Corpino, R.; Ricci, P.C.
Mater. Sci. Eng.,C 2003, 23, 1073-1076.
[16]
Liu, Z.C.; Chen, H.R.; Huang, W.M.; Gu, J.L.; Bu, W.B.; Hua, Z.L.;
Shi, J.L. Microporous Mesoporous Mater. 2006, 89, 270-275.
[17]
Sokolov, I.; Volkov, D.O. J. Mater. Chem. 2010, 20, 4247-4250.
[18]
Dujardin, E.; Mann, S. Adv. Mater. 2002, 14,11, 775-788.
[19]
Wu, L.-Q.; Payne, G.F. Trends Biotechnol. 2004, 22, 593-599.
[20]
Losic, D.; Mitchell, J.G.; Voelcker, N.H. Adv. Mater. 2009, 21, 29472958.
11
[21]
Yang, W.; Lopez, P.J.; Rosengarten, G.R. Analyst 2011, 136, 42-53.
[22]
Rusmini, F.; Zhong, Z.; Feijen, J. Biomacromolecules 2007, 8, 17751789.
[23]
Starodub, N.F.; Fedorenko, L.L.; Starodub, V.M.; Dikij, S.P.;
Svechnikov, S.V. Sens. Actuators, B 1996, 35-36, 44-47.
[24]
Di Francia, G.; LaFerrara, V.; Manzo, S.; Chiavarini, S. Biosens.
Bioelectron. 2005, 21, 661-665.
[25]
Huang, J.-G; Lee, C.-L.; Lin, H.-M.; Chuang, T.-L.; Wang, W.-S.;
Juang, R.-H.; Wang, C.-H.; Lee, C.-K.; Lin, S.-M.; Lin, C.-W. Biosens.
Bioelectron. 2006, 22, 519-525.
[26]
Chaniotakis, N.; Sofikiti, N. Anal. Chim. Acta 2008, 615, 1-9.
[27]
Jianrong, C.; Yuqing, M.; Nongyue, H.; Xiaohua, W.; Sijiao, L.
Biotechnol. Adv. 2004, 22, 505-518.
[28]
Jane, A.; Dronov, R.; Hodges, A.; Voelcker, N.H. Trends Biotechnol.
2009, 27, 230-239.
[29]
Vo-Dinh, T.; Cullum, B. J. Anal. Chem. 2000, 366, 540-551.
[30]
Marazuela, M.D.; Moreno-Bondi, M.C. Anal. Bioanal. Chem. 2002,
372-664-682.
[31]
Schaferling, M.; Nagl, S. Anal. Bioanal. Chem. 2006, 385, 500-517.
[32]
Zhang, S. Nat. Biotechnol. 2003, 21, 10, 1171-1178.
[33]
Sarikaya, M.; Tamerler, C.; Schwarz, D.T.; Baneyx, F. Annu. Rev.
Mater. Res. 2004, 34, 373-408.
[34]
Scala, S.; Bowler, C. Cell. Mol. Life Sci. 2001, 58-1666-1673.
[35]
Morse, D.E. Trends Biotechnol. 1999, 17, 230-232.
[36]
Round, F.E. Nova Hedwagia 1972, 23:449-63.
[37]
Crawford, S.A.; Higgins, M.J.; Mulvaney, P.; Wetherbee, R. J. Phycol.
2001, 37, 543-554.
12
[38]
The Biology of Diatoms, edited by D. Werner, Botanical Monographs,
University of California Press, Berkeley, CA (1977).
[39]
Brzezinski, M.A.; Olson, R.J.; Chisholm, S.W. Mar. Ecol. Prog. Ser.
1990, 67, 83-96.
[40]
Sumper, M.; Kroger, N. J. Mater. Chem. 2004, 14, 20-59-2065.
[41]
Hildebrand, M.; Frigeri, L.G.; Davis, A.K. J. Phycol. 2007, 43, 730740.
[42]
Kroger, N.; Deutzmann, R.; Sumper, M. Science 1999, 286,5 11291132.
[43]
Brzezinski, M.A.; Conley, D.J. J. Phycol. 1994, 30, 45-55.
[44]
Conway, H.L.; Harrison P.J. Mar. Biol. 1977, 43, 33-43.
[45]
Conway, H.L.; Harrison, P.J.; Davis, C.O. Mar. Biol. 1976, 35, 187199.
[46]
Thamatrakoln, K.; Hildebrand, M. Plant Physiol. 2008, 146, 13971407.
[47]
De Stefano, L.; Lamberti, A.; Rotiroti, L.; De Stefano, M. Acta
Biomater. 2008, 4, 126-130.
[48]
De Stefano, L.; Rotiroti, L.; De Stefano, M.; Lamberti, A.; Lettieri, S.;
Setaro, A.; Maddalena, P. Biosens. Bioelectron. 2009, 24,1580-1584.
[49]
Townley, H.E.; Parker, A.R.; White-Cooper, H. Adv. Funct. Mater.
2008, 18, 369-374.
[50]
De Stefano, L.; Rendina, I.; De Stefano, M.; Bismuto, A.; Maddalena,
P. Appl. Phys. Lett. 2005, 87, 233902.
[51]
Setaro, A.; Lettieri, S.; Maddalena, P.; De Stefano, L. Appl. Phys. Lett.
2007, 91, 051921.
[52]
Lettieri, S.; Setaro, A.; De Stefano, L.; De Stefano, M.; Maddalena, P.
Adv. Funct. Mater. 2008, 18, 1257-1264.
13
[53]
Qin, T.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. J. Nanosci.
Nanotechnol. 2008, 8, 1-7.
[54]
Butcher, K.S.A.; Ferris, J.M.; Phillips, M.R.; Fouquet-Wintrebert, M.;
Jong Wah, J.W.; Jovanovic, N.; Vyverman, W.; Chepurnov, V.A.
Mater. Sci. Eng., C 2005, 25, 658-663.
[55]
Carbonaro, C.M.; Clemente, F.; Corpino, R.; Ricci, P.C.; Anedda, A. J.
Phys. Chem. B 2005, 109, 14441-1444.
[56]
Glinka, L.D.; Lin, S.-H.; Chen, Y.T. Phys. Rev. B: Condes. Matter.
2002, 66, 035404-1-10.
[57]
Rückschloss, M.; Wirschem. Th.; Tamura, H.; Ruhl, G.; Oswald, J.;
Veprek, S. J. Lumin. 1995, 63, 279-287.
[58]
Lauerhaas, J.M.; Sailor, M.J. Science 1993, 261, 1567-1568.
[59]
Vaccaro, L.; Cannas, M.; Boscaino, R. Solid State Commun. 2008, 146,
14-151.
[60]
Tsybeskov, L.; Vandyshev, J.V.; Fauchet, P.M. Phys. Rev. B. Condes.
Matter 1994, 49, 7821-7824.
[61]
[62]
Lockwood, D.J.; J. Mater. Sci.: Mater. Electron. 2009, 20, S235–S244.
Gao, T.; Bao, X.M.; Yan, F.; Tong, S. Phys. Lett. A 1997, 232, 321325.
[63]
Shen, J.K.; Wu, X.L. Bao, X.M.; Yuan, R.K.; Zou, J.P.; Tan, C. Phys
Lett. A 2000, 273, 208-211.
[64]
Li, J.; Wu, X.L.; Yang, Y.M.; Yang, X.; Bao, X.M. Phys. Lett. A 2003,
314, 299-303.
[65]
[66]
Molle, A.; Bhuiyan, M.N.K.; Tallarida, G.; Fanciulli, M. Mater. Sci.
Semicond. Process. 2006, 9, 673–678.
Neustrev, V.B. J. Phys. Condens. Matt. 1994, 6, 6901-6936.
[67]
Alessi, A.; Agnello, S.; Gelardi, F.M.; Grandi, S.; Magistris, A.;
Boscaino, R. Opt. Express 2008, 16, 7, 4895-4900.
[68]
Lewin, J. Phycologia 1966, 6, 1-12.
14
[69]
Azam, F.; Hemmingsen, B.B.; Volcani, B.E. Arch. Mikrobiol. 1973,
92, 11-20.
[70]
Chiappino, M.L.; Azam, F.; Volcani, B.E. Protoplasma 1977, 93, 191204.
[71]
Thomas, W.H.; Dodson, A.N. Mar. Biol. 1974, 27, 11-19.
[72]
Jeffryes, C.; Gutu, T.; Jiao, J.; Rorrer, G.L. Mater. Sci. Eng. C 2008,
28, 107-118.
[73]
Qin, T.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. ACS Nano 2008,
2, 1296-1304.
[74]
Jeffryes, C; Solanki
[75]
Messina, F.; Agnello, S.; Boscaino, R.; Cannas, M.; Grandi, S.;
Quartarone, E. J. Non-Crys. Solids 2007, 353, 670-673.
15
(b)
(a)
25 µm
1 µm
(c)
5 µm
Figure 1-1. Scanning electron microscope images of diatom frustules. (a)
Cyclotella sp. (b) Coscinodiscus wailessi (c) Pinnularia sp.
16
Parent diatom
cell
S
Hypovalve
deposited
Hypovalve SDV
formed
G2 + M
G1
Girdle band
lengthened
Girdle band
fabrication
Daughter cell
separation
Figure 1-2. Schematic of the diatom cell division cycle and frustule
formation.
17
Chapter 2
Photoluminescence Detection of Biomolecules by Antibody-Functionalized
Diatom Biosilica
Debra K. Galea, Timothy Gutub, Jun Jiaob, Chih-hung Changa, and Gregory L.
Rorrera
[a]
D.K. Gale, C.-H. Chang, G.L. Rorrer
Department of Chemical Engineering, Oregon State University,
Corvallis, Oregon 97331, USA
[b]
T. Gutu, J. Jiao
Department of Physics, Portland State University, Portland, Oregon
97207, USA
Published in Advanced Functional Materials
2009, 19, 926-933
18
Abstract
Diatoms are single-celled algae that make microscale silica shells called
“frustules”, which possess intricate nanoscale features imbedded within
periodic two-dimensional pore arrays. In this study, antibody-functionalized
diatom biosilica frustules serve as a microscale biosensor platform for selective
and label-free photoluminescence (PL)-based detection of immunocomplex
formation. The model antibody rabbit immunoglobulin G (IgG) is covalently
attached to the frustule biosilica of the disk-shaped, 10-µm diatom Cyclotella
sp. by silanol amination and crosslinking steps to a surface site density of
3948±499 IgG molecules µm-2. Functionalization of the diatom biosilica with
the nucleophilic IgG antibody amplifies the intrinsic blue PL of diatom
biosilica by a factor of six. Furthermore, immunocomplex formation with the
complimentary antigen anti-rabbit IgG further increases the peak PL intensity
by at least a factor of three, whereas a non-complimentary antigen (goat antihuman IgG) does not. The nucleophilic immunocomplex increases the PL
intensity by donating electrons to non-radiative defect sites on the
photoluminescent diatom biosilica, thereby decreasing non-radiative electron
decay and increasing radiative emission. This unique enhancement in PL
emission is correlated to the antigen (goat anti-rabbit IgG) concentration,
where immunocomplex binding follows a Langmuir isotherm with binding
constant of 2.8±0.7 x 10-7M.
19
Introduction
Antibody-antigen immunocomplex formation is an essential tool to
diagnose and profile immune diseases rapidly and reliably [1]. The
immunocomplex is typically detected and quantified by optical, electrical, or
mass measurement methods using surface-immobilized biomolecules [2-4].
Electrical and mass measurement methods are often destructive, invasive, and
yield low signal intensities. Furthermore, the labels used in many optical
methods can introduce contamination to the sample matrix. Biosensing
platforms that use label-free optical detection, such as ellipsometry and surface
plasmon resonance, avoid these limitations. However, ellipsometry can only
sense biomolecules on a flat surface of small surface area, and it is difficult to
detect multiple biomolecules with surface plasmon resonance [4].
Several studies have shown that the photoluminescent properties of
porous silicon and nanoscale semiconductor materials change upon their
interaction with biomolecules [5-14]. This phenomenon may serve as the basis
for the label-free optical detection of immunocomplex formation that does not
require fluorescence-based resonance energy transfer (FRET).
However, no
previous studies have harnessed enhanced photoluminescence resulting from
the binding of an antigen with a nucleophilic, antibody-functionalized
photoluminescent surface to selectively report immunocomplex formation.
In this study, we exploit the PL properties of diatom biosilica to detect
the binding of a complimentary antigen with an antibody-functionalized
diatom frustule. Diatoms are single-celled algae which fabricate intricately
patterned biosilica shells called “frustules” that possess a surface rich in
reactive silanol groups. Unlike mesoporous silicates, centric diatom frustules
assume a microscale, disk-like form factor with a flat surface possessing a
periodic pore structure ordered at the nanoscale (Figure 1). These structures
are particularly attractive for bionanotechnology applications, given the current
20
interest in new approaches for biomolecule nanopatterning [15-18]. Recent
studies have shown that methods for the covalent attachment of antibodies to
silicate materials can be extended to diatom biosilica [19,20] with an eye
towards microscale “lab-on-chip” applications. Furthermore, nanostructured
diatom biosilica possesses blue photoluminescence which can be controlled
through cell cultivation [21-23]. The intrinsic photoluminescence property of
non-functionalized diatom biosilica has already been harnessed for gas sensing
applications [24-27], but not for biosensing.
Below, we describe our chemistry for the functionalization of intact
biosilica frustules isolated from cultured Cyclotella cells with the model
antibody Rabbit Immunoglobulin G (IgG). The distribution of antibodies on
the frustule surface is quantitatively profiled by transmission electron
microscopy through immunocomplex formation with its gold nanoparticlelabeled complimentary antigen (anti-Rabbit IgG). The selectivity of antigen
detection and the mechanisms underlying enhanced photoluminescence are
discussed. Finally, we correlate enhanced PL emission to antigen
concentration in solution to illustrate the potential of antibody-functionalized
diatom frustules for selective, label-free biosensing applications.
Results
Fine Features of Cyclotella Frustule Biosilica. The biosilica frustule
of the centric diatom Cyclotella sp. consists of two adjoined halves, each
consisting of a disc-shaped valve resting on top of a ring-shaped girdle band.
Isolation of the frustule from the cell wall of Cyclotella sp. diatom cells by
aqueous hydrogen peroxide treatment generally separated the valve and girdle
band structures. SEM and TEM images of a representative valve are presented
in Figure 1. The frustule valve has an ornately patterned surface ordered at the
submicron and nanoscales. The pore structure consists of ~200 nm pores that
21
ring the perimeter of the valve (rimportulae), and linear arrays of ~100 nm
diameter pores that project from the center of the valve to the rim. Lining the
base of each ~100 nm pore is a thin layer of silica containing four to five
nanopores of ~20 nm diameter.
Characterization and Imaging of Antibody-Functionalized
Frustule Biosilica. The frustule biosilica surface was functionalized with the
model antibody Rabbit Immunoglobulin G (IgG) by the two-step reaction
process described in the Experimental section. The basic structure of the
antibody-functionalized diatom biosilica is presented in Figure 2.
In the first step, the surface silanol (≡Si-OH) groups on the diatom
biosilica were functionalized with amine groups by reaction with 3aminopropyltrimethoxysilane (APS).
Fourier transform infrared (FT-IR) spectra of bare diatom biosilica and
amine-functionalized diatom biosilica are compared in Figure 3. FT-IR spectra
clearly showed characteristic peaks for diatom biosilica, including for Si-O-Si
bending at 470 and 800 cm-1 [28], Si-O stretching of Si-OH groups at 950 cm-1
[28], Si-O-Si stretching at 1095 cm-1 [29], and O-H stretching of surfacebound hydroxyl groups at 3435 cm-1, which would include bound water and
Si-O-H. Also present were characteristic peaks for carboxyl (C=O) stretching
at 1735 cm-1 [29] as well as C-H stretching at 2850 (CH2) and 2925 (CH3) cm-1
[29]. Therefore, residual organic materials still adhered to the biosilica
frustule after hydrogen peroxide treatment. After amine functionalization, the
FTIR spectra had a strong peak at 1390 cm-1, which is characteristic for C-H
bending of the Si-O-CH3 moiety [30]. Furthermore, the C-H stretching peaks
at 2850 and 2925 cm-1 were considerably enhanced, which were assigned to
the propyl group on the amine-functionalized biosilica. The weaker peaks at
1525 and 1640 cm-1 were assigned to the N-H bending and stretching of the
22
amine group [29,30], and the peak near 3435 cm-1 was enhanced, which was
also assigned to N-H stretching [30].
The free amine (-NH2) sites anchored to the diatom biosilica were
detected by reactive labeling with fluorescamine to yield a fluorophor which
emitted blue-green light at 464 nm when excited at 380 nm under epifluorescent microcope imaging (Figure 4a). The amine groups were evenly
distributed across the face of the valve, with a higher density observed along
the rim, commensurate with the thickened solid biosilica surrounding the
rimportulae. Bare diatom biosilica valves treated with fluorescamine had no
fluorescence.
In the second step of the functionalization process, the Rabbit IgG
antibody was covalently bound to the amine-functionalized diatom biosilica by
the BS3 crosslinking reaction [31]. When challenged with Goat anti-Rabbit
IgG labeled with DyLightTM 488, the entire surface of the antibodyfunctionalized valve emitted green light at 518 nm when excited at 493 nm
under epi-fluorescent microcope imaging (Figure 4b). Two control
experiments validated the specificity of antigen binding. First, bare diatom
biosilica challenged with Goat anti-Rabbit IgG labeled with DyLight 488 had
no emission under epifluorescent microscope imaging. Second, Rabbit-IgG
antibody-functionalized diatom biosilica challenged with Goat anti-Human
IgG labeled with FITC, a non-complimentary antigen, also had no emission.
The density of Rabbit IgG molecules on the antibody-functionalized
valve surface was estimated by immunocomplex formation with Goat antiRabbit IgG that was labeled with 10 nm gold nanoparticles. In order to test for
specificity, diatom biosilica, amine-functionalized diatom biosilica, and
Rabbit-IgG antibody-functionalized biosilica were challenged with gold
nanoparticle labeled Goat anti-Rabbit IgG. The TEM images presented in
Figures 4c and 4d show the gold nanoparticles as ~10 nm black dots uniformly
23
distributed across the valve surface. The composition of the nanoparticles was
confirmed as elemental Au by X-ray dispersive analysis. By statistical
sampling and image analysis of several valves (5 valves per sample, three 0.25
2
images per valve, 4.5 antigen molecules per gold nanoparticle), the
density of antibody molecules on the valve surface was estimated, as shown in
Figure 5. From comparison of Figures 4c and 4d, it was also apparent that
some of the gold nanoparticle-labeled Goat anti-Rabbit IgG molecules were
nonspecifically bound to the amine-functionalized diatom biosilica. Westcott
et al. [32] also reported that gold nanoparticles can bind to free NH2 functional
groups on amine-functionalized 100 nm silica nanoparticles.
Enhanced Photoluminescence of Antibody-Functionalized Diatom
Biosilica for Selective and Quantitative Complimentary Antigen
Detection. Representative photoluminescence (PL) spectra of bare diatom
biosilica, Rabbit IgG antibody-functionalized diatom biosilica, and antibodyfunctionalized diatom biosilica bound to its complimentary antigen (Goat antiRabbit IgG) are compared in Figure 6. The bare diatom biosilica possessed
intrinsic blue photoluminescence at 337 nm excitation. Therefore, the peak PL
spectra were normalized to the peak PL intensity of the bare diatom biosilica,
as shown in Figure 7a. Functionalization of the diatom biosilica with Rabbit
IgG enhanced the PL intensity by a factor of over six relative to the bare
diatom biosilica, but did not change the shape of the spectra, which peaked in
the blue range of 430-450 nm. The peak PL intensity of the Goat anti-Rabbit
IgG bound to Rabbit IgG antibody-functionalized diatom biosilica was 2.6
times higher than the peak PL intensity of antibody-functionalized diatom
biosilica alone, and over 16 times that of bare diatom biosilica.
Two control experiments validated the specificity of the enhanced PL
emission associated with the immunocomplex of the antibody-functionalized
diatom frustule biosilica with its complimentary antigen. First, when Rabbit
24
IgG antibody-functionalized diatom biosilica was challenged with the noncomplimentary antigen Goat anti-Human IgG (Figure 7a), there was no
statistically significant change in PL intensity relative to the antibodyfunctionalized diatom biosilica alone (ANOVA, p = 0.42 > α0.05). Second,
when Rabbit IgG in solution was physically mixed with bare diatom biosilica,
centrifuged, and then washed again, there was no change in the PL spectra
relative to wet diatom biosilica alone. Furthermore, Rabbit IgG and Goat antiRabbit IgG dissolved in PBS buffer possessed had no PL emission at 337 nm
excitation.
A third control experiment demonstrated that the enhanced PL emission
required diatom biosilica as the functionalization platform. A thermally
grown SiO2 thin film on a silicon wafer substrate was animated with APS and
then functionalized with Rabbit IgG by the BS3 crosslinking reaction. Amine
groups on the thermally grown SiO2 thin film were validated with the amine
reactive fluorescent probe fluorescamine, and Rabbit IgG molecules on the
SiO2 thin film were validated following binding with FITC-labeled anti-Rabbit
IgG (data not shown). The peak PL intensities of the antibody-functionalized
SiO2 film before and after binding to its complimentary antigen are compared
in Figure 7b. The PL intensity of the thermally grown SiO2 thin film
decreased after antibody functionalization and binding of the antibodyfunctionalized surface with its complimentary antigen. This control
experiment result was consistent with the earlier work of Starodub et al. [5],
who observed decreased photoluminescence of an IgG immunocomplex
adsorbed on silicon.
The effect of antigen (Goat anti-Rabbit IgG) concentration on peak PL
emission from the antibody-functionalized diatom biosilica is presented in
Figure 8. In these experiments, antibody-functionalized diatom biosilica
deposited on a glass slide was incubated with the antigen solution under gentle
25
mixing until equilibrium was achieved. The molar ratio of antigen in solution
to antibody on the diatom biosilica was in significant excess relative to that
required for saturation binding, so that the antigen concentration in solution did
not decrease significantly following its partitioning onto the solid.
The data
were fit to a Langmuir adsorption isotherm model of the form
I C
I  I o  max A (1)
K A  CA
where CA is the concentration of anti-Rabbit IgG (mol/L), KA is the affinity
constant for Rabbit IgG/Goat anti-Rabbit IgG binding (mol/L), I is the peak PL
intensity (counts), Imax is the saturation peak PL intensity (counts), and Io is the
PL intensity of the Rabbit IgG functionalized biosilica itself (Io = 6,157 ±
1,380 counts). The fitted parameters, estimated by nonlinear regression
(Marquardt method, Statgraphics), were KA = 2.8 ± 0.710-7 mol/L and Imax =
14,558 ± 1,074 counts. All errors are reported as 1.0 standard error (S.E.) at
95% confidence.
Discussion
Below, the unique features of the antibody-functionalized diatom
frustules for selective and quantitative detection of biomolecules and the likely
origins of photoluminescence emission enhancement upon immunocomplex
formation are discussed.
The antibody-functionalized diatom frustule surface of Cyclotella sp.
has unique topographical features for bio-nanotechnology applications in
general and biosensing in particular. The valve surface is corrugated at the
nanoscale, as the solid surface between the ordered pore arrays is defined by
peaks and valleys (Figure 1), and the surface itself is composed of silica
nanoparticles [33]. In immunosensing applications, nano-corrugated surfaces
26
are known to increase the extent of antigen capture and sensitivity of antigen
detection [34-37].
Recently, methods for the covalent attachment of antibodies to silicate
materials have been extended to diatom biosilica isolated from the large (~200
µm) centric diatom Coscinodiscus. In a preliminary study, De Stefano et al.
[19] used glutaraldehyde to attach murine monoclonal antibody (MoAb) to
amine-derivatized diatom biosilica, and validated the presence of MoAb by
complimentary binding with a rhodamine-labeled G23 peptide under
fluorescent imaging. Townley et al. [20] used the UV-activated crosslinking
reagent N-5-azido-2-nitrobenzoyloxysuccinimide (ANB-NOS) to attach the
carbohydrate moiety on the antibody Immunoglobulin Y (Ig-Y) to aminederivatized diatom biosilica. Furthermore, Townley et al. extended this
process to detect multiple antibodies by FITC labeling. However, both of
these studies did not attempt to quantify the extent of antibody
functionalization on the diatom biosilica.
In this study, we validated both amine derivatization and antibody
Cyclotella by FT-IR and epi-fluorescent labeling (Figures 3, 4a, 4b).
Furthermore, we quantitatively estimated the antibody site density and
nanoscale distribution on the frustule surface by TEM imaging of the goldnanoparticle labeled immunocomplex (Figures 4c, 4d). The estimated site
density on the antibody(Figure 5). The patterning of the immunocomplex reflected the patterning on
the valve biosilica surface. Assuming a 1:1 antigen to antibody ratio, this
corresponded to a nominal antibody spacing of 22.5 nm, assuming both sides
of the valve were uniformly functionalized. In comparison, the dimensions of
the Y-shaped IgG molecule are 14.2 nm across and 8.5 nm base height [38].
2
27
This is the first reported study of using enhanced photoluminescence
(PL) to selectively and quantitatively detect immunocomplex formation on an
antibody-functionalized biosilica surface. Previous work in our laboratory has
shown that diatom biosilica possesses intrinsic photoluminescence that is
dependent upon the development of nanoscale features on the frustule surface
during the diatom cell cultivation [ 21-23]. Studies of De Stefano and coworkers have shown that the adsorption of electrophilic and nucleophilic gas
molecules onto diatom biosilica changed the PL spectra [24-27]. In general,
photoluminescence emission is generated when electron-hole pairs created by
absorbing incident photons from the excitation source radiatively recombine
and discharge photons of lower energy. Although bulk SiO2 is an insulator
with a wide band gap of ~11 eV [39], Si-O surface defects associated with
nanoporous silica create energy states that give rise to visible blue-green
photoluminescence at 400-500 nm (3.1-2.5 eV) under UV excitation [40-42].
Nucleophilic moieties or biomolecules that are attached to nanoscale
semiconductors or other photoluminescent surfaces with nanoscale topology
can increase PL emission [43,6,44, 45,7-14]. Many of these studies proposed
that nucleophilic groups increase the PL intensity by donating electrons to nonradiative defect sites on the photoluminescent surface, thereby decreasing nonradiative electron decay and increasing the radiative emission, resulting in a
higher quantum efficiency of the functionalized photoluminescent surface. In
this study, the PL intensity increased over six times after the diatom biosilica
was covalently functionalized with the Rabbit IgG antibody (Figure 6), a result
consistent with this mechanism since Rabbit IgG is nucleophilic [46]. When
the Rabbit IgG-functionalized diatom biosilica was bound to its complimentary
antigen (Goat anti-Rabbit IgG), which is also nucleophilic, the peak PL
intensity increased again by at least a factor of two. Furthermore, the PL
intensity did not change when the Rabbit-IgG functionalized diatom biosilica
28
was challenged the non-complimentary antigen anti-Human IgG, which
demonstrated that the PL detection was selective (Figure 7a). A thermally
grown SiO2 thin film, which lacks the nanoscale topology of diatom biosilica,
did not exhibit enhanced photoluminescence when functionalized with Rabbit
IgG (Figure 7b).
The selective increase in PL intensity upon immunocomplex formation
on the antibody-functionalized diatom biosilica offers a label-free method for
biomolecule detection. The increase in PL intensity was dependent on the
antigen concentration in solution (Figure 8), which illustrates the potential for
quantitative biosensing applications of this novel phenomenon. The increase
in PL intensity with antigen concentration was described by a simple Langmuir
model for immunocomplex formation. The binding constant estimated from
the PL emission vs. antigen concentration data, KA = 2.8 ± 0.710-7 M, was
within the range of binding constants for the Rabbit IgG/anti-Rabbit IgG
immunocomplex estimated by other detection methods [47-49].
The fabrication and performance evaluation of a biosensor device was
beyond the scope of this study. However, the antibody-functionalized
Cyclotella frustule surface is its own microscale (~10 µm) biosensing platform,
where the diatom biosilica serves as the transducer to selectively report antigen
binding and its concentration through enhanced photoluminescence. This
microscale biosensor platform could be integrated into a biosensor device in a
variety of ways. For example, diatom frustules, each functionalized with a
separate antibody, could be deposited as a monolayer in defined patterns by
microfluidic printing [50] to facilitate simultaneous, multi-antigen detection.
Alternatively, each disk-shaped antibody-functionalized frustule could be
mounted onto the tip of a fiber optic cable to excite the sample and collect the
photoluminescence spectra. Towards this end, Losic et al. [51] developed a
29
technique to attach a single diatom frustule valve onto the tip of a glass
capillary tube.
Conclusions
Biosilica frustules isolated from the cell wall of the centric diatom
Cyclotella sp. (Figure 1) were functionalized with the model antibody Rabbit
IgG to yield a uniform, nanopatterned antibody surface that selectively binds to
its complimentary antigen, Goat anti-Rabbit IgG (Figures 2-5).
Functionalization of the diatom biosilica with the nucleophilic IgG antibody
amplified the intrinsic blue photoluminescence of diatom biosilica (Figure 6).
Furthermore, the Rabbit IgG immunocomplex increased the peak PL intensity
by at least a factor of two, whereas a non-complimentary antigen (Goat antiHuman IgG) did not (Figure 7). We proposed that the nucleophilic
immunocomplex increased the PL intensity by donating electrons to nonradiative defect sites on the photoluminescent diatom biosilica, thereby
decreasing non-radiative electron decay and increasing the radiative emission.
This novel phenomenon was used to correlate PL emission to antigen (Goat
anti-Rabbit IgG) concentration (Figure 8), where immunocomplex binding
followed a Langmuir isotherm with binding constant of KA = 2.8 ± 0.710-7
mol/L. These results suggest that antibody-functionalized diatom biosilica
frustules may serve as a novel microscale biosensor platform for the selective,
label-free, photoluminescence-based detection of immunocomplex formation.
Experimental
Diatom Cell Culture. Axenic cultures of the photosynthetic marine
diatom Cyclotella sp. were obtained from UTEX The Culture Collection of
Algae (UTEX # 1269) and cultivated in 500 mL non-agitated flasks in 100 mL
of Harrison’s Artificial Seawater Medium enriched with f/2 nutrients and 0.7
30
mM silicon as Na2SiO3 [52]. Cultures were maintained at 22 ºC and 170 μE m2 -1
s incident light intensity on a 14 h light, 10 h dark photoperiod. The cell
suspension was subcultured at 10% v/v every 14 days. Intact biosilica
frustules were isolated from 14 day old diatom cells by treatment with 30 wt%
aqueous hydrogen peroxide at pH 2.5 as described by Jeffryes et al. [53], and
stored in methanol.
Amine Functionalization of Diatom Biosilica. Surface silanol (≡SiOH) groups on the diatom biosilica were functionalized with amine groups by
reaction with 3-aminopropyl-trimethoxysilane (APS). Specifically, a 10 mg
aliquot of diatom biosilica, 2.0 mL anhydrous ethanol, and 50 μL of APS
(Sigma-Aldrich # 09326, mol. Wt. 179.29) were mixed together, heated to
80ºC in an amber glass vial under constant mixing for 2 h, and then reacted
unstirred for an additional 24 h [32]. The suspension was centrifuged (10 min,
325 g) and washed in ethanol three times to remove excess APS. The final
amine-functionalized diatom biosilica was stored in ethanol.
Covalent Attachment of Antibody to Amine-functionalized Diatom
Biosilica. To facilitate photoluminescence spectroscopy measurements,
antibody functionalization was performed on a thin film deposit of aminefunctionalized diatom biosilica. A 20 μL aliquot of 1.0 mg/mL of aminefunctionalized diatom biosilica suspended in deionized water was pipetted onto
an 18 mm circular glass slide and then dried in air. A second 20 μL aliquot was
deposited and dried to prepare a 5 mm diameter film of 0.040 mg of aminefunctionalized diatom biosilica. The film was oven dried in air for 1.0 h at 90
o
C. Rabbit Immunoglobulin G (IgG) antibody molecules were covalently
attached to the amine-functionalized diatom biosilica thin film by the BS3
crosslinking reaction, using methods adapted from Hermanson [31]. First, the
BS3 crosslinking reagent was covalently attached to free –NH2 groups on the
amine-functionalized diatom biosilica. Specifically, a single amine-
31
functionalized biosilica thin film was transferred to a single well of a
polystyrene six well plate which contained 1.8 mL of PBS buffer (phosphate
buffered saline at pH 7.2). A 200 µL aliquot of a 0.0657 mg/mL BS3 solution
(Bis(sulfosuccinimidyl)suberate, Pierce Biotechnology # 21580, mol. Wt.
572.43, 0.574 µmol BS3/g amine-functionalized biosilica) was added then
added to the well. The well plate was mixed on an orbital shaker at 60 rpm at
room temperature for 20 min. Next, the Rabbit IgG antibody was covalently
coupled to the BS3 crosslink on the amine-functionalized biosilica. The
biosilica thin film was dip-rinsed in PBS buffer and then transferred to a new
well containing 1.8 mL of PBS buffer and 200 µL of a 0.0234 mg/mL Rabbit
IgG solution (ImmunoPure Rabbit IgG, Pierce Biotechnology # 31235, mol.
Wt. ~150,000, 7.8 10-4 µmol IgG/g amine-functionalized biosilica). The well
plate was mixed on an orbital shaker at 60 rpm at room temperature for 2.0 h.
The antibody-functionalized biosilica thin film was then rinsed in PBS buffer.
The diatom biosilica firmly adhered to the glass slide during all preparation
steps. All samples were prepared in triplicate.
Binding of Complimentary Antigen to Antibody-Functionalized
Diatom Biosilica. Rabbit IgG-functionalized diatom biosilica thin films were
challenged with different concentrations of its complimentary antigen, Goat
anti-Rabbit IgG. After binding, the immunocomplexed thin films were
analyzed by photoluminescence (PL) spectroscopy. In these experiments, a
single antibody-functionalized biosilica thin film described above was
transferred to a single well of a polystyrene six-well plate containing 2.0 mL
PBS buffer at pH 7.2. A stock solution of ImmunoPure Goat anti-Rabbit IgG
(Pierce Biotechnology #31210) was added to the well to provide the desired
antigen concentration, and the well plate was mixed on an orbital shaker at 60
rpm at room temperature for 2.0 h.
The anti-Rabbit IgG concentrations in the
well plate ranged from 8.010-5 mg/mL (5.310-10 mol/L) to 2.4 mg/mL (1.610-
32
6
mol/L). After binding, the immunocomplexed diatom thin films were dip-
rinsed in PBS buffer to remove excess antigen, dip-rinsed twice in deionized
water to remove salts, and then dried in air at room temperature prior to PL
measurement. Each binding experiment at a given antigen concentration was
performed in triplicate.
Immunocomplex Specificity. To test for immunocomplex specificity,
the Rabbit IgG antibody-functionalized diatom frustule biosilica was
challenged with complimentary Goat anti-Rabbit IgG and non-complimentary
Goat anti-Human IgG. A large stoichiometric excess of a given antigen was
added to Rabbit IgG antibody-functionalized diatom biosilica powder
suspended in PBS buffer at pH 7.2, and then mixed on a tube rotator at 8 rpm
for 2 h. The suspension was centrifuged and washed three times in 1.0 mL PBS
buffer to remove excess antigen. Three forms of Goat anti-Rabbit IgG antigen
were tested: ImmunoPure Goat anti-Rabbit IgG (Pierce Biotechnology
#31210) for photoluminescence measurements, DyLightTM 488 labeled Goat
anti-Rabbit IgG (Pierce Biotechnology #35552) for antibody-functionalized
diatom valve imaging by epi-fluorescent microscopy, and 10 nm gold
nanoparticle labeled Goat anti-Rabbit IgG (Sigma-Aldrich #G7402, 4.5 IgG
per gold nanoparticle) for TEM imaging and estimation of antibodyfunctionalized site density. Two forms of anti-Human IgG were also tested:
ImmunoPure Goat anti-Human IgG (H+L) (Pierce Biotechnology #31130) for
photoluminescence measurements, and FITC labeled Goat anti-Human IgG
(Sigma-Aldrich # F3512) for fluorescence imaging.
Antibody Functionalization of Thermally Grown SiO2 Thin Film.
A silicon wafer was cut with a diamond knife into 5 x 5mm pieces and then
annealed in air for 3 h at 1000ºC within a preheated furnace to prepare
thermally grown SiO2 thin film coupons. The thermally grown SiO2 thin film
33
was then functionalized with Rabbit IgG as described above for diatom
biosilica.
FT-IR Spectroscopy. Fourier transform infrared (FT-IR)
spectroscopic analysis was performed on bare diatom biosilica and aminefunctionalized diatom biosilica. A transparent pellet for FT-IR analysis was
prepared by grinding 2.0 mg of an oven-dried (80 oC, 4 h) diatom frustule
sample with 100 mg KBr with a mortar and pestle. The powder compressed
for 1.0 min and then mounted onto the FT-IR sample holder. FT-IR spectra
were collected with a Nicolet 510P FT-IR spectrophotometer equipped with
DTGS detector, averaging 32 scans with a resolution of 4 cm-1.
Epifluorescence Microscopy. Epi-fluorescent microscopy was used to
image the distribution of free –NH2 groups on the amine-functionalized diatom
biosilica valve and the distribution of fluorescently labeled Goat anti-Rabbit
IgG bound to the Rabbit IgG antibody-functionalized valve. Aminefunctionalized sites were detected with the fluorescent probe fluorescamine
(FluoroPureTM grade, Molecular Probes # F-20261), which reacts with the
amine group [54]. Specifically, 10 µL of 0.5 mg/mL fluorescamine dissolved
in methanol was added to a 1.0 mL suspension of amine-functionalized diatom
frustule biosilica diluted in ethanol (0.01 mg diatom/mL) within a black
polypropylene microcentrifuge tube. The mixture was vortexed for 2 min to
complete the reaction. Fluorescent images were acquired at 100X
magnification and 30 sec exposure time under a 380 nm excitation source,
using a Leica DM inverted light microscope equipped with a DAPI longpass
filter (425 nm). Similarly, fluorescent images of Rabbit IgG antibodyfunctionalized valves bound to DyLight 488 labeled anti-Rabbit IgG antigen
were acquired under a 470 nm excitation source and GFP filter (525 nm).
Rabbit IgG antibody functionalized valves challenged with FITC-labeled anti-
34
Human IgG antigen were imaged under 470 nm excitation source and GFP
filter (525 nm).
Electron Microscopy. Bare diatom frustules and Rabbit-IgG
antibody-functionalized diatom frustules bound to 10 nm gold nanoparticle
labeled Goat anti-Rabbit IgG were imaged by scanning transmission
microscopy (SEM), transmission electron microscopy (TEM), and scanning
transmission microscopy equipped with x-ray dispersion spectroscopy (STEMEDS). A given diatom frustule suspension was centrifuged, re-suspended in
methanol, deposited onto a holey carbon coated copper TEM grid, and allowed
to dry in air. Frustules were then imaged by TEM or STEM-EDS at 200 keV
with an FEI Tecnai F20 TEM. Frustules were also deposited onto carbon tape
then imaged on a FEI Sirion field emission SEM at 1.5 keV.
Photoluminescence Spectroscopy. The optical emission properties of
functionalized diatom biosilica were characterized by photoluminescence (PL)
spectroscopy. The immunocomplexed diatom biosilica thin films prepared
above were excited with a 337 nm N2 gas laser source (Spectra Physics VSL,
30 kW peak power, 2.4 mW average power, 20 Hz, 4 nsec pulse, 120 µJ). The
laser beam was cut to a 1 by 3 mm slit and aimed at the sample on a 45o angle.
The fraction of light emitted from the sample surface at a 45o angle was sent
through a 360 nm UV cut-off filter to remove the reflected laser excitation
signal, focused to a 1.0 mm beam width, and then measured with an Acton
Inspectrum 300 spectrometer equipped with CCD detector (0.20 mm slit width,
300 gratings/mm, 2000 msec integration time). Functionalized diatom
biosilica from the immunocomplex specificity experiments was centrifuged
(10 min, 325 g), and 1.0 mg of wet pellet was loaded into a PL sample holder
described previously [22]. These samples were excited with a 337 nm light
source from a 175 W Xenon lamp (Spectral Products, # ASB-XE-175EX)
equipped with a Spectral Products CM100-1/8 monochromator (1200
35
grooves/mm, 0.60 mm slit width) with a 400 nm cut-off filter. The PL
spectrum on the same diatom biosilica sample was measured after selected
steps in functionalization process (bare diatom biosilica, antibodyfunctionalized diatom biosilica, immunocomplexed diatom biosilica) to
facilitate PL data normalization to bare diatom biosilica. All PL measurements
were carried out at room temperature and corrected for the baseline spectra.
Control experiments showed that the residual PBS buffer had no effect on the
PL spectra. All PL measurements were performed in triplicate.
Acknowledgements
This research was supported by the National Science Foundation (NSF) under
Nanoscale Interdisciplinary Research Team (NIRT) award number BES0400648.
References
[1]
F. Rusmini, Z. Zhong, J. Feijen, Biomacromolecules 2007, 8, 1775-1789.
[2]
T. Vo-Dinh, B. Cullum, Fresenius J. Anal. Chem. 2000, 366, 540-551.
[3]
M.D. Marazuela, M.C. Moreno-Bondi, Anal. Bioanal. Chem. 2002, 372,
664-682.
[4]
M. Schäferling, S. Nagl, Anal. Bioanal. Chem. 2006, 385, 500-517.
[5]
N.F. Starodub, L.L. Fedorenko, V.M. Starodub, S.P. Dikij, S.V.
Svechnikov, Sensor Actuat B-Chem. 1996, 35-36, 44-47.
[6]
M.P. Stewart, J.M. Buriak, Adv. Mater. 2000, 12, 859-869.
[7]
T.-Y. Liu, H.-C. Liao, C.-C. Lin, S.-H. Hu, S.-Y. Chen, Langmuir 2006,
22, 5804-5809.
[8]
Q. Wang, Y. Kuo, Y. Wang, G. Shin, C. Ruengruglikit, Q. Huang, J.
Phys. Chem. B. 2006, 110, 16860-16866.
[9]
A. Dorfman, N. Kumar, J. Hahm, Adv. Mater. 2006, 18, 2685-2690.
36
[10] A. Dorfman, N. Kumar, J. Hahm, Langmuir 2006, 22, 4890-4895.
[11] S.N. Sarangi, K. Goswami, S.N. Sahu, Biosens. Bioelectron. 2007, 22,
3086-3091.
[12] Q. Shang, H. Wang, H. Yu, G. Shan, R. Yan, Colloid Surface A 2007,
294, 86-91.
[13] J.-H. Wang, H.-Q. Wang, H.-L. Zhang, X.-Q. Li, X.-F. Hua, Z.-L.
Huang, Y.-D. Zhao, Colloid Surface A 2007, 305, 48-53.
[14] M. Dybiec, G. Chornokur, S. Ostapenko, A. Wolcott, J.Z. Zhang, A.
Zajac, C. Phelan, T. Sellers, D. Gerion, Appl. Phys. Lett. 2007, 90,
263112.
[15] K.L. Christman, V.D. Enriquez-Rios, H.D. Maynard, Soft Matter 2006,
2, 928-939.
[16] T. Blättler, C. Huwiler, M. Ochsner, B. Städler, H. Solak, J. Vörös, H.M.,
Grandin, J. Nanosci. Nanotechnol. 2006, 6, 2237-2264.
[17] P.M. Mendes, C.L. Yeung, J.A. Preece, Nanoscale Res. Lett. 2007, 2,
373-384.
[18] D. Erickson, S. Mandal, A. H.J. Yang, B. Cordovez, Microfluid.
Nanofluid. 2008, 4, 33-52.
[19] L. De Stefano, A. Lamberti, L. Rotiroti, M. De Stefano, Acta Biomater.
2008, 4, 126-130.
[20] H.E. Townley, A.R. Parker, H. White-Cooper, Adv. Funct. Mater. 2008,
18, 369-374.
[21] C. Jeffryes, R. Solanki, Y. Rangineni, W. Wang, C.-H. Chang, G.L.
Rorrer, Adv. Mater. 2008, 20, 2633-2637.
[22] T. Qin, T. Gutu, J. Jiao, C.-H. Chang, G.L. Rorrer, J. Nanosci.
Nanotechnol. 2008, 8, 2392-2398.
[23] T. Qin, T. Gutu, J. Jiao, C.-H. Chang, G.L. Rorrer, ACS Nano 2008, 2,
1296-1304.
37
[24] L. De Stefano, I. Rendina, M. De Stefano, A. Bismuto, P. Maddalena,
Appl. Phys. Lett. 2005, 87, 233902.
[25] A. Setaro, S. Lettieri, P. Maddalena, L. De Stefano, Appl. Phys. Lett.
2007, 91, 051921.
[26] A. Bismuto, A. Setaro, P. Maddalena, L. De Stefano, M. De Stefano,
Sensor. Actuat. B-Chem. 2008, 130, 396-399.
[27] S. Lettieri, A. Setaro, L. De Stefano, M. De Stefano, P. Maddalena, Adv.
Funct. Mater. 2008, 18, 1257-1264.
[28] A. Gendron-Badou, T. Coradin, J. Maquet, F. Fröhlich, J. Livage, J. NonCryst. Solids 2003, 316, 331-337.
[29] A. Gélabert, O.S. Pokrovsky, J. Schott, A. Boudou, A. Feurtet-Mazel, J.
Mielczarski, E. Mielczarski, N. Mesmer-Dudons, O. Spalla, Geochim.
Cosmochim. Acta 2004, 68, 4039-4058.
[30] T. Kovalchuk, H. Sfihi, L. Kostenko, V. Zaitsev, J. Fraissard, J. Colloid
Interface Sci. 2006, 302, 214-229.
[31] G.T. Hermanson, Bioconjugate Techniques, Academic Press, London,
1996.
[32] S.L. Westcott, S.J. Oldenberg, T.R. Lee, N.J. Halas, Langmuir 1998, 14,
5396-5401.
[33] S.A. Crawford, M.J. Higgins, P. Mulvaney, R. Wetherbee, J. Phycol.
2001, 37, 543-554.
[34] K. Borchers, A. Weber, H. Brunner, G.E.M. Tovar, Anal. Bioanal. Chem.
2005, 383, 738-746.
[35] C. Wang, F.L. Yap, Y. Zhang, Colloid Surface B 2005, 46, 255-260.
[36] P.V. Tuttle, A.E. Rundell, T.J. Webster, Int. J. Nanomed. 2006, 1, 497505.
[37] L.M. Bonanno, L.A. DeLouise, Langmuir 2007, 23, 5817-5823.
38
[38] A. Nisonoff, J.E. Hopper, S.B. Spring, The Antibody Molecule,
Academic Press, New York, 1975.
[39] H. Ibach, J.E. Rowe, Phys. Rev. B 1974, 10, 710-718.
[40] Y.D. Glinka, A.S. Zyubin, A.M. Mebel, S.H. Lin, L.P. Hwang, Y.T.
Chen, Chem. Phys. Lett. 2002, 358, 180-186 .
[41] J.M. Shieh, A.-T. Cho, Y.-F. Lai, B.-T. Dai, F.-M. Pan, K.-J. Chao,
Electrochem. Solid-State Lett. 2004, 7, G319-G322.
[42] C.M. Carbonaro, F. Clemente, R. Corpino, P.C. Ricci, A. Anedda, J.
Phys. Chem. B 2005, 109, 14441–14444.
[43] T. Dannhauser, M. O’Neil, K. Johansson, D. Whitten, G. McLendon, J.
Phys. Chem. 1986, 90, 6074-6076.
[44] L.D. Carlos, R.A. Sá Ferreira, R.N. Pereira, M. Assunção, V. De Zea
Bermudez, J. Phys. Chem. B 2004, 108, 14924-14932.
[45] R.P. Jia, Y. Shen, H.Q. Luo, X.G. Chen, Z.D. Hu, D.S. Xue, Thin Solid
Films 2005, 471, 264-269.
[46] S. Paul, Y. Nishiyama, S. Planque, S. Karle, H. Taguchi, C. Hanson,
M.E. Weksler, Springer Semin. Immun. 2005, 26, 485-503.
[47] J.J. Gooding, C. Wasiowych, D. Barnett, D.B. Hibbert, J.N. Barisci, G.G.
Wallace, Biosens. Bioelectron. 2004, 20, 260-268.
[48] E. Briand, M. Salmain, C. Compère, C.-M. Pradier, Biosens. Bioelectron.
2007, 22, 2884-2890.
[49] D.A. Gish, F. Nsiah, M.T. McDermott, M.J. Brett, Anal. Chem. 2007, 79,
4228-4232.
[50] P. Calvert, Chem. Mater. 2001, 13, 3299-3305.
[51] D. Losic, G. Rosengarten, J.G. Mitchell, N.H. Voelcker, J. Nanosci.
Nanotechnol. 2006, 6, 982-989.
[52] P.J. Harrison, R.E. Waters, F.J.R. Taylor, J. Phycol. 1980, 16, 28-35.
39
[53] C. Jeffryes, T. Gutu, J. Jiao, G.L. Rorrer, Mater. Sci. Eng. C 2008, 28,
107-118.
[54] G. Deng, M.A. Markowitz, P.R. Kust, B.P. Gaber, Mat. Sci. Eng. C 2000,
11, 165-172.
40
Figure 2-1. Electron microscopy images of a representative biosilica frustule
valve isolated from the centric diatom Cyclotella sp. (a) SEM image of whole
valve, revealing surface topology;
(b) TEM image of whole valve, revealing radial symmetry of the nanopore
arrays; (c) TEM image of nanopore pattern detail.
41
BS3 crosslink
APS-biosilica
IgG- anti-IgG
CH3
Si
O
O
H
Si
N
O
N
O
O
H
CH3
Figure 2-2. The basic structure of the antibody-functionalized diatom biosilica
(not drawn to scale).
1095
950
800
470
1735
1640
1525
1390
2925
2850
3435
42
50
diatom biosilica
40
30
20
amine-functionalized
diatom biosilica
10
0
4000 3400 2800 2200 1600 1000
400
Wavenumber (cm-1)
Figure 2-3. FT-IR spectra of frustule biosilica before and after amine
functionalization.
43
(b)
(a)
5 mm
5 mm
(c)
(d)
(c)
(d)
100 nm
100 nm
Figure 2-4. Imaging of functionalized diatom biosilica. (a) Epifluorescent
image of amine-functionalized diatom frustule labeled with fluorescamine; (b)
Epifluorescent image of Rabbit IgG antibody-functionalized diatom frustule
labeled with DyLightTM 488 Goat anti-Rabbit IgG.; (c) TEM image of aminefunctionalized diatom frustule challenged with Goat anti-Rabbit IgG labeled 10
nm gold nanoparticles (4.5 antigens/gold nanoparticle); (d) TEM image of
antibody-functionalized diatom frustule challenged with Goat anti-Rabbit IgG
labeled 10 nm gold nanoparticles (4.5 antigens/gold nanoparticle).
Antigens on diatom surface ( # / m m 2) .
44
5000
4000
3000
2000
1000
0
diatom
biosilica
diatom
NH2-biosilica
diatom +
antibody
Figure 2-5. Comparison of surface site density of Goat anti-Rabbit IgG on
the valve of bare diatom biosilica, amine-functionalized diatom biosilica
(diatom NH2-biosilica), and Rabbit IgG antibody functionalized diatom
biosilica (diatom+antibody).
45
18
Normalized PL intensity
16
diatom +
antibody +
antigen
14
12
10
8
diatom +
antibody
6
4
2
0
diatom
-2
350
400
450
500
550
600
Wavelength (nm)
Figure 2-6. Comparison of photoluminescence (PL) spectra of bare diatom
biosilica, Rabbit IgG antibody-functionalized diatom biosilica
(diatom+antibody), and Rabbit IgG immunocomplex functionalized diatom
biosilica (diatom+antibody+antigen). Excitation wavelength was 337 nm.
Inset: photograph of PL emission from antibody-functionalized diatom
biosilica.
46
Normalized PL peak intensity
20
434
3 nm
15
449
10
444
15 nm
11 nm
5
445
6 nm
0
diatom
frustule
diatom +
antibody
diatom +
diatom +
antibody + antibody +
non-comp.
antigen
antigen
Normalized PL Peak Intensity
1.50
1.25
401
1 nm
1.00
401
1 nm
0.75
0.50
403
3 nm
0.25
0.00
silicon dioxide silicon dioxide + silicon dioxide +
antibody
antibody +
antigen
Figure 2-7. Comparison of normalized peak PL intensity of antibody
functionalized diatom biosilica with that of an antibody-functionalized,
thermally grown SiO2 thin film. (a) Bare diatom biosilica, Rabbit IgG
antibody-functionalized diatom biosilica (diatom+antibody), Rabbit IgGfunctionalized diatom biosilica with Goat anti-Rabbit IgG
(diatom+antibody+antigen), and Rabbit IgG antibody-functionalized diatom
biosilica challenged with a non-complimentary Goat anti-Human IgG
(diatom+antibody+non-comp. antigen). (b) Thermally grown SiO2, Rabbit
IgG antibody-functionalized SiO2 (SiO2 +antibody), and Rabbit IgG
immunocomplex functionalized SiO2 (SiO2+antibody+antigen). Excitation
wavelength was 337 nm.
Photoluminescence Intensity, I (counts)
47
25,000
20,000
15,000
10,000
KA= 2.8 x 10-7 M
5,000
0
1.0E-10
1.0E-08
1.0E-06
Antigen Concentration, CA (M)
1.0E-04
Figure 2-8. Peak PL intensity of Rabbit IgG antibody functionalized biosilica
vs. Goat anti-Rabbit IgG antigen concentration. Solid line presents best fit to
Equation (1).
48
Figure 2-9. This work was featured on the inside cover of Advanced
Functional Materials, volume 19, number 6, 2009.
49
Chapter 3
Thermal annealing activates amplified photoluminescence of germanium
metabolically doped in diatom biosilica
Debra K. Gale,*a Clayton Jeffryes, a Timothy Gutu,b Jun Jiao, b Chih-hung
Chang a and Gregory L. Rorrer a
[a]
D.K. Gale, C.-H. Chang, G.L. Rorrer
Department of Chemical Engineering, Oregon State University,
Corvallis, Oregon 97331, USA
[b]
T. Gutu, J. Jiao
Department of Physics, Portland State University, Portland, Oregon
97207, USA
Published in the Journal of Materials Chemistry
2011, 21, 10658-10665
50
Abstract
There is significant interest in the fabrication of germanium (Ge) doped silica
for optoelectronic device applications. In this study, highly 50hotoluminescent
Ge centers, metabolically doped into diatom biosilica, are activated by thermal
annealing in air. Diatoms are single celled photosynthetic algae that make
silica shells called frustules. These frustules possess intricate features and
patterns on the nano- and micro-scale. A two stage photobioreactor cultivation
strategy is used to biologically fabricate diatom biosilica doped with Ge,
ranging from 0.24 to 0.96 weight percent Ge. X-ray photoelectron
spectroscopy (XPS) and electron diffraction show that a mixture of amorphous
germanium dioxide (GeO2) and germanium oxide (GeO) is doped into the
frustule structure. Annealing in air thermally converts the amorphous GeO2 to
GeO, commensurate with an enhancement in the photoluminescence. Thermal
gravimetric analysis (TGA) and photoluminescence of annealed biosilica with
and without Ge, confirms that the photoluminescence originates from GeO
50hotoluminescent centers, and not from the inherent photoluminescence of
the biosilica. This is the first study to thermally activate and characterize
highly 50hotoluminescent Ge centers metabolically doped into diatom
biosilica.
51
Introduction
Nanostructured silica has received significant attention due to its visible
photoluminescence.[1] Recently there has been interest in doping silica with
germanium (Ge), which has a higher carrier mobility. [2] Nanostructured silica
doped with Ge exhibits enhanced optoelectronic properties important for
microelectronic device applications such as the next generation of display
devices. Current methods of fabrication to dope silica with Ge employ top
down, low throughput processes such as laser ablation, [3] ion implantation, [4]
and co-sputtering, [2a, 2b,2c, 5] which require extremes of temperature, pressure,
and power. Furthermore, with increased interest in incorporating these
nanostructured semiconductor materials into miniaturized devices for
promising optoelectronic applications, equipment and cost of production is
challenged. [6]
To address the limitations of conventional semiconductor material
fabrication, there is an emergence of interest in bio-fabrication techniques to
assemble nano- to micro-scale hierarchical semiconductor materials. [7]
Specifically, diatoms have been touted as a paradigm for biological fabrication
of nanostructured silica. [8] Diatoms, single celled algae, fabricate silica shells
also known as frustules, which possess nanoscale patterned pore arrays within
a microstructure morphology. These are desirable features for a semiconductor
material. Recently, significant efforts have been made to harness the inherent
optoelectronic properties of nanostructured diatom frustules to use as a
platform for device applications. [9] The photoluminescence of biosilica
frustules has been used for immunocomplex sensing, [10] antibody detection,
[11]
and gas sensing. [12] The biosilica has also been chemically converted to
BaTiO3 and SrTiO3, [13] and coated in ZnSiO4:Mn, Zn2SiO4, TiO2,
Y2SiO5:Eu3+, CdS, and gold, [14] all in an effort to enhance the optoelectronic
52
properties.
In this study, we show how thermal activation of Ge doped in diatom
biosilica creates highly 52hotoluminescent frustules which are uniquely
patterned on the nano- and micro-scale. It is advantageous to use diatoms to
make 52hotoluminescent Ge doped silica as opposed to industrial production
methods because these structures are made biologically, without extremes of
temperature, pressure, power and sophisticated equipment. Furthermore,
diatom biosilica has uniquely patterned submicron and nanoscale features
within a microscale form factor, which is difficult to achieve with current
fabrication methods.
Experimental Section
Two-Stage Diatom Cell Cultivation: Pure cultures of the photosynthetic
marine diatom Pinnularia sp. Were obtained from UTEX Culture Collection of
Algae (UTEX # B679). Maintenance of this culture in the laboratory has been
previously described. [15a] Germanium was metabolically doped into the
biosilica structure with a two-stage photobioreactor process. [15a] Pinnularia
sp. Was cultivated in bubble-column photobioreactors under 150 µ E m-2 s-1
incident light intensity, 14 hour light/10 hour dark photoperiod, 1.0 L air L-1
culture min-1 aeration rate, and 22ºC. In stage I, Pinnularia sp. Cells (1.0x105
± 1.6x104 cells mL-1) were inoculated into the photobioreactor with sodium
metasilicate, Si(OH)4, (0.78 mM) and grown to the point of Si depletion and
the final target cell density (5.25x105 ± 6.5x104 cells mL-1). In stage II, a
mixture of Si(OH)4 (0.53 mM Si) and Ge(OH)4 (4.25, 17.3, 25.2, or 38.3 µM
Ge) was fed to the Si starved cells. These initial Ge concentrations in the cell
culture medium in stage II yielded biosilica doped with Ge (0.243±0.081,
0.452±0.072, 0.498±0.061, and 0.965±0.012 weight percent), as measured by
ion-coupled plasma (ICP) analysis using a Varian (Liberty 150) ICP emission
53
spectrometer.[15a]
Frustule Isolation: Pinnularia sp. Frustules were isolated from the diatom
cells for electron microscopy, photoluminescence (PL) spectroscopy, and
elemental analysis by ICP. Diatoms were treated with aqueous hydrogen
peroxide (30 weight percent, pH 2.5) to oxidize organic cell components and
remove carbonates, followed by gentle washing in deionized water, then
suspension in MeOH to yield intact isolated frustules.[15a, 10]
Electron Microscopy: Frustules suspended in MeOH were deposited on
holey carbon copper grids and imaged by transmission electron microscopy
(TEM) or scanning transmission electron microscopy- energy dispersive X-ray
spectroscopy (STEM-EDX) with an FEI Tecnai F20 high resolution TEM at
200 keV.
Thermal Annealing: Biosilica frustules, isolated from the diatom cells, were
thermally annealed in air for 2.0 hours at 400°C within a preheated furnace.
The PL was measured before and after annealing of the same material. The PL
was not measured until the samples were allowed to cool for 1 hour to reach
room temperature.
Photoluminescence Spectroscopy: An aliquot of diatom frustule suspension
(60 µL of 1 mg mL-1) in deionized water was deposited onto a circular glass
coverslip and allowed to air dry. This resulted in a thin layer of frustule
powder material (5mm in diameter) for photoluminescence spectroscopy
measurements. The frustules were excited with a 337 nm N2 gas laser source
(Spectra Physics VSL, 30 kW peak power, 2.4 mW average power, 20 Hz, 4 ns
pulse, 120 µJ). The frustule photoluminescence was sent through a 360 nm
UV cut off filter and measured with an Acton Inspectrum 300 spectrometer
equipped with a CCD detector (0.2 mm slit width, 300 gratings per
53hotolumin, 2000 ms integration time), which was described in our previous
work. [10] All PL measurements were carried out in triplicate.
54
Thermal Gravimetric Analysis: A TA Instruments Thermogravimetric
Analyzer 2950 was used for thermogravimetric analysis (TGA). Pinnularia
sp. Frustules were put under vacuum conditions for 7 days prior to TGA. A
frustule mass (1.5 mg) was ramped from 22ºC to 150ºC at 3ºC min-1 in air,
then held at 150ºC for 120 minutes then ramped from 150ºC to 600ºC at 3ºC
min-1. TGA measurements were carried out in triplicate.
X-Ray Photoelectron Spectroscopy: The biosilica frustule mass (60 µg) used
for PL was also used for X-ray Photoelectron Spectroscopy (XPS), except it
was deposited on a cleaned silicon
wafer coupon to minimize sample charging. Biosilica powder was 54hotolum
for the germanium oxide chemical structure with
the Ge2p electron using a Thermo Scientific K-Alpha XPS with an Al Kα xray source (1486.6 eV photons, 200 eV pass energy, 100 ms dwell time, 0.200
eV energy step size, 400 µm spot size, 300 scans) using an Ar flood gun for
charge compensation. Spectral deconvolution was performed by Avantage 4.1
software using the Shirley method of background subtraction, a FWHM upper
limit of 3.5 nm and a combination of Gaussian and Lorentzian peak shapes.
Spectra were corrected for energy drift due to charge compensation with the
O1s electron of SiO2 at 532 eV as an internal standard. [17a]
Results
Metabolic Insertion of Germanium Defects into the Diatom Biosilica
Ge was metabolically inserted into the nanostructure of the Pinnularia sp.
Diatom biosilica using a two-stage photobioreactor cultivation strategy. [15] In
Stage I of the photobioreactor cultivation, a diatom cell suspension was
inoculated into a 4.0 L volume of LDM and seawater medium to a cell number
density of 1.0x105 ± 1.6x104 cells mL-1 and supplemented with 0.78 mM
sodium metasilicate, speciated to Si(OH)4, to allow for a five fold increase in
55
the cell number density. The end of Stage I was
identified by silicon depletion in the culture medium, commensurate with
silicon starvation by the cells, and a constant cell number density for two
photoperiods. A representative cell number density and silicon concentration
profile in the cell culture medium in Stages I and II is presented in Figure 1a.
Stage II was initiated by the addition of a mixture if Si(OH)4 and Ge(OH)4 to
yield a final target concentration of 0.53 mM Si and 4.25, 17.3, 25.2, or 38 µM
Ge in the cell culture medium. The concentration of Si was designed to allow
for one cell doubling in Stage II and the concentration of Ge was varied to
control the concentration of Ge in the biosilica. A representative Si and Ge
concentration profile is shown in Figure 1b. The Si and Ge were rapidly
depleted from the culture medium by a surge uptake mechanism into the
diatom cells within a 10 hour period. [16]
At the end of Stage II, the diatom cells were treated with hydrogen peroxide
to oxidize organic cell components and isolate the inorganic frustule biosilica
for analysis of elemental Ge and Si in the biosilica measured by ICP, as
described in the experimental section. Initial concentrations of Ge in the
culture medium of 4.25, 17.3, 25.2, and 38.3 µM Ge yielded final Ge weight
percent of 0.243±0.081, 0.452±0.072, 0.498±0.061, and 0.965±0.012 Ge in the
biosilica.
Electron Microscopy of Diatom Frustules
Transmission electron microscopy (TEM) images were acquired to capture the
submicron and nanopore morphology of the biosilica frustules. TEM images
were acquired from control cultivation experiments at the end of Stage II,
where no germanium was added to the culture medium. A TEM image of a
representative Pinnularia sp. Frustule obtained at the end of Stage II of the
cultivation is shown in Figure 2a and Figure 2b. Diatom frustules consist of a
56
upper and lower valve, the epitheca and hypotheca, respectively, which are
connected with a thickened strip of silica called the girdle band. The
Pinnularia sp. Diatom frustule is approximately 30 µm in length. The frustule
has submicron features, Figure 2c, in the form of a linear array of pores, which
are approximately 200 nm, which are filled with nanoscale patterned pores on
the order of 50 nm.
TEM images of representative Pinnularia sp. Frustules doped with 0.49 wt
% Ge and annealed at 400ºC for 2 hours in air are presented in Figure 3. TEM
images show the Pinnularia sp. Diatom microstructure (Figure 3a, Figure 3b),
submicron structure (Figure 3c, Figure 3d), and the nanostructure (Figure 3e,
Figure 3f). Metabolic insertion of Ge defects into the diatom frustule appeared
to fuse the 50 nm pores into larger 200 nm pores. As a result, the submicron
structure of the frustule appeared as fused slits.
Thermal Annealing of Diatom Biosilica and Electron Diffraction
The diatom biosilica was thermally annealed to activate Ge 56hotoluminescent
centers in the frustule. Biosilica, isolated by the hydrogen peroxide method,
was annealed in air for 2.0 hours at 400ºC within a pre-heated furnace. The
crystalline structure of the biosilica, before and after annealing, was evaluated
by electron diffraction. The inset in Figure 3f is representative of an electron
diffraction pattern acquired from a frustule containing 0.49 weight percent Ge,
which was annealed in air for 2.0 hours at 400ºC. The Ge doped biosilica
before and after thermal annealing exhibited an amorphous structure.
Activation of Ge Photoluminescence by Thermal Annealing
The photoluminescence (PL) of the biosilica with and without germanium was
characterized as a function of annealing temperature. A 60 µg layer of diatom
biosilica powder was deposited on a circular glass coverslip and allowed to air
57
dry. [10] The PL of the biosilica was measured, then annealed for 2 hours in air
at a desired temperature, followed by a post-annealed PL measurement of the
same material using a 337 nm N2 excitation source. The PL of biosilica, which
contained 0.45 weight percent Ge, annealed at temperatures up to 600ºC, is
shown in Figure 4 a. The PL signal intensity is normalized to the PL signal
intensity of the biosilica before thermal annealing treatment which is
represented as 1.0 in Figure 4a. The PL intensity of the
Ge doped biosilica was enhanced when annealed between 150ºC and 400°C,
reaching a maximum enhancement of 3.5 times at an annealing temperature of
250ºC, when compared to the PL of the Ge doped biosilica before annealing.
Between 500ºC and 600ºC, the PL of the Ge doped biosilica was quenched to
approximately half that of the room temperature PL, but was still enhanced
compared to the native biosilica without Ge. The PL, as a function of
annealing temperature, was measured for biosilica without Ge which was
generated from the control cultivation at the end of Stage II, where no Ge was
added to the culture medium. The PL of the native biosilica, without Ge,
remained unchanged when annealed at 150ºC, but was completely quenched
for all annealing temperatures above 200ºC, as shown in Figure 4b.
Thermal gravimetric analysis (TGA) of the biosilica was used to characterize
the removal of physisorbed water and 57hotoluminescent surface silanols on
the biosilica as a function of temperature. The temperature of the biosilica was
ramped from 22ºC to 150ºC at 3ºC per minute and maintained at 150ºC for 2
hours, followed by a temperature ramp of 3ºC per minute to 600ºC. A
representative TGA profile of biosilica is shown in Figure 4c. The weight loss
percent from the TGA is presented in Figure 4a and Figure 4b, next to the
complimentary PL as a function of annealing temperature.
The PL of biosilica doped with 0.24, 0.45, 0.49, and 0.96 weight percent Ge,
which was annealed at 250ºC, 400ºC, and 600ºC is presented in Figure 5a.
58
Biosilica doped with at least 0.45 weight percent Ge exhibited the most intense
enhancement after thermal annealing at 250ºC. This biosilica also exhibited a
PL enhancement after annealing at 400ºC. However, after annealing at 600ºC,
the biosilica PL was quenched for all samples, despite the Ge dopant
concentration. The biosilica with the lowest level of Ge, 0.24 weight percent,
behaved similarly to the biosilica without Ge, and exhibited quenched PL after
annealing at all temperatures. The PL emission was in the blue range between
430 nm to 460 nm for all Ge dopant concentrations and annealing
temperatures, except for the biosilica without Ge and the biosilica with 0.24
weight percent Ge, which shifted red to almost 560 nm when annealed at
250ºC. The PL spectrum of native biosilica (increased by a factor of 5 for
scale) and biosilica doped with 0.96 weight percent Ge, which was annealed at
250ºC, is presented in Figure 6. The peak intensity wavelength of the biosilica
is 576 nm and 456 nm for native biosilica and biosilica doped with 0.96 weight
percent Ge. Photographs of the biosilica powder excited with the 337 nm laser
is shown in the Figure 6 inset.
Germanium Oxide Photoluminescent Defects
X-ray photoelectron spectroscopy (XPS) was used to 58hotolu the presence of
germanium oxide (GeO) and germanium dioxide (GeO2) 58hotoluminescent
centers doped into the biosilica. Biosilica was 58hotolum for the germanium
oxide chemical structure with the Ge2p electron using a Thermo Scientific KAlpha XPS with an 58hotolumi x-ray source. The Ge2p electron spectrum
was deconvoluted into the GeO and GeO2 signal contributions. Figure 7a
shows the evolution of GeO2 and GeO in biosilica doped with 0.96 weight
percent Ge, as a function of annealing temperature, which was normalized to
highest signal intensity. Figure 7b shows the XPS peak height intensity ratio
of GeO to GeO2 at each annealing temperature. GeO2 was identified at a
59
binding energy of 1220.2 eV, by measuring the spectrum of industrial GeO2,
and GeO was identified at a binding energy of 1221.6 eV. [17] The Ge doped
biosilica which was as deposited (not annealed), yielded the lowest signal
intensity for GeO and GeO2 which were approximately of the same intensity.
Upon annealing at 250ºC the GeO and GeO2 signal almost doubled in
intensity, but remained at the same relative concentration. However, the Ge
signal for biosilica, which was annealed at 400ºC, was comprised of an intense
GeO peak with a very weak GeO2 signal. The GeO and GeO2 signals for the
biosilica annealed at 600ºC exhibit weaker signal intensities and are similar to
the signal intensity of the as deposited biosilica. This data confirmed that a
mixture of GeO2 and GeO was doped into the biosilica frustule by the cell, and
the oxide form was selectively altered by annealing.
Discussion
This study has shown that living diatom cells can be used as a platform to
biologically fabricate highly 59hotoluminescent Si and Ge metal oxide nanoand micro-structured materials. This is a novel, environmentally benign
method to biologically fabricate nanostructured Si and Ge metal oxide
materials, which is an attractive alternative method to conventional
semiconductor fabrication techniques.
Germanium 59hotoluminescent centers were metabolically doped into the
frustule biosilica of the diatom Pinnularia sp. Using a two stage
photobioreactor cultivation strategy. Previously we showed that this
cultivation strategy uniformly doped Ge into the nano- and micro-structure of
the diatom frustule.[15a] This study is unique from previous work because it
characterizes the chemical structure of Ge doped into the biosilica and the
mechanism of PL enhancement activated by thermal annealing in air.
Silicon is a required substrate for frustule formation and cell division. When
60
the diatom cell is in a silicon starved state, by Si depletion in the culture
medium and a constant cell density, the cell will take up Si and Ge, ultimately
doping the nano- and micro-structure of the biosilica frustule with Ge, shown
in Figure 1. [15a, 18] The diatom frustule is comprised of an upper and lower
valve, held together with a girdle band. Upon division, the diatom cell
fabricates two new daughter valves. Both of the new daughter cells have one
parent valve made up of silica and one daughter valve made up of Ge doped
silica. As a result, statistically only half of the frustule valves will be doped
with Ge. Since the elemental Ge analysis was a bulk measurement, the local
Ge content doped into the daughter valve is actually twice the reported value
which was measured by ICP. Representative TEM images of frustules doped
with 0.49 weight percent Ge, which were annealed at 400ºC are presented in
Figure 3. Metabolic doping of Ge into the biosilica structure induced aberrant
micro, submicron, and nano structures. There were not any Ge rich particles
observed with TEM before or after annealing.
Comparison of the PL of native biosilica and Ge doped biosilica presented
Figure 4, Figure 5, and Figure 6 is strong evidence that GeO acts as a
60hotoluminescent center in the frustule biosilica. The PL of biosilica without
Ge was quenched after annealing. Whereas the biosilica doped with at least
0.45 weight % Ge was enhanced after annealing at 250ºC and 400ºC. The blue
PL of diatom biosilica up to 500 nm, originates from surface defects as silanol
(SiOH) groups and the green PL up to 600 nm, originates from ≡Si-H, shown
in Figure 5b and Figure 6, [19, 10] like that of nanostructured silica. [1b, 1d, 20]
Since the PL of the native biosilica is quenched by annealing in air, and the PL
of the Ge doped biosilica is enhanced by annealing in air, this suggests that the
enhanced PL originates entirely from Ge defects. Like the bio-fabricated
silica, Ge doped silica made by conventional semiconductor fabrication
techniques also exhibits blue photoluminescence, which is attributed to Ge
61
defects.[2a, 4a, 2c, 5c] In addition to annealing temperature, the PL of the biosilica
can be controlled by the level of Ge defects doped into the biosilica (Figure
5a). The concentration of Ge in the biosilica tuned the PL enhancement of the
biosilica after it was annealed, which is further evidence that the PL is a
function of Ge defects metabolically doped into the biosilica.
TGA of the biosilica confirmed that surface silanol groups, which are a
source of diatom photoluminescence, were removed by thermal annealing.
Two significant weight losses were observed from the TGA of the native
biosilica and Ge doped biosilica in Figure 4c, one at 150ºC and a second by
400ºC. Thermal annealing in air at 150ºC will remove physisorbed water, and
at 400ºC condensed silanol groups from the silica surface. [21, 20c, 20d] It was
also shown on diatom biosilica that silanols begin to condense on the surface at
200ºC.[22] Although the total weight loss shown by the TGA varied from 15
% to 30 %, between 150ºC and 400ºC, for the Ge doped biosilica and native
biosilica, this does not represent the loss of only surface silanols. Diatom
frustules are known to contain proteins interstitially located within the
biosilica. [23]
In addition to surface silanols, the weight loss after 150ºC is
likely proteins located within the biosilica. The SiOH 61hotoluminescent
centers were selectively condensed from the biosilica surface by thermal
annealing in air to remove the contribution of the silica to the total
photoluminescence signal, to expose the PL of the Ge centers. The PL of the
native biosilica was quenched by 75% by annealing at 250ºC, which suggests
that the majority of silanol groups were removed by this annealing
temperature. Commensurate with the quenched PL of the native biosilica at
250ºC, the Ge doped biosilica PL was enhanced by 3.5 times the pre-annealed
biosilica PL, where it remained enhanced compared to the native biosilica for
all annealing temperatures. This TGA data shows that the surface silanols,
which are a source of the biosilica PL, can be removed by thermal annealing to
62
expose the PL from the Ge defects. Figure 6 compares the PL spectrum of
native biosilica and biosilica with 0.96 weight % Ge which was annealed at
250ºC. After thermal annealing at 250ºC, the peak intensity wavelength of the
biosilica shifts from approximately 450 nm to 570 nm, which is characteristic
of PL originating from ≡Si-H surface defects. [1b, 1d, 20] Although the
62hotoluminescent surface silanols have been removed by thermal annealing,
the total PL of the Ge doped biosilica is comprised of a strong contribution
from Ge defects and a very weak contribution from ≡Si-H surface defects. The
emission energy is visibly evident in the photograph of biosilica and Ge doped
biosilica annealed at 250ºC presented in the Figure 6 inset. The biosilica emits
a light green color, whereas the Ge doped biosilica exhibits a bright blue
emission.
The final line of evidence which supports that 62hotoluminescent Ge centers
doped into diatom biosilica were thermally activated, was the X-ray
photoelectron spectroscopy (XPS) analysis. A mixture of germanium oxide
(GeO) and germanium dioxide (GeO2) was doped into the frustule biosilica by
the diatom cell as shown by XPS analysis of the Ge2p electron in Figure 7,
where 1220.2 eV and 1221.6 eV are the binding energies for GeO2 and GeO.
[17]
The concentration of GeO and GeO2 was increased in equal proportions
after annealing at 250ºC, consistent with an enhancement in the PL.
Furthermore, after annealing at 400ºC, the concentration of GeO was over
twice that of GeO2, which also exhibited an enhanced PL compared to the PL
of the native biosilica. Like Ge doped biosilica, Ge implanted into SiO2 show
enhanced blue PL as a function of increased annealing temperature up to 500ºC
and increase in GeO concentration, shown by XPS. [2a, 5b, 3b] The mechanism of
the blue PL in Ge implanted silica, as with Ge doped biosilica we biologically
fabricated, originates from GeO 62hotoluminescent centers. [5a, 4a, 2c, 3] GeO2
implanted into silica will begin to thermally decompose into GeO between
63
400ºC and 500ºC, [2b, 2d, 24] which explains the increase in GeO concentration in
the biosilica after annealing at 400ºC. Above temperatures of 500ºC, the GeO
will likely begin to thermally desorb from the surface, and the concentrations
of germanium oxides will decrease, as shown in Figure 7, after the biosilica is
annealed at 600ºC. [24c, 25]
The conditions which produce biosilica which emits the most intense PL
originating from GeO centers, is biosilica doped with 0.96 wt. % Ge and
annealed at 400 °C, evident by Figure 7b. At an annealing temperature of
250°C, the total PL signal originates from germanium oxide and the intrinsic
PL of the biosilica. Furthermore, at an annealing temperature of 600°C, the PL
from the biosilica and the GeO has been thermally quenched. The PL signal of
the Ge doped biosilica annealed at 400°C originates solely from the GeO
centers because the PL originating from the biosilica has been thermally
quenched.
Conclusions
In this study, we showed that the PL of biosilica metabolically doped with Ge
can be thermally activated to create a highly 63hotoluminescent nano- and
micro-structured material. Controlled levels of Ge were doped into the diatom
biosilica frustule by a two-stage photobioreactor cultivation strategy.
Germanium oxide 63hotoluminescent defects were enhanced by thermally
converting the GeO2 deposited into the frustule by the diatom cell into highly
63hotoluminescent GeO. Nanostructured Si-Ge metal oxide composite
materials made by diatom cells can be biologically fabricated on a large scale
under environmentally benign conditions. These materials not only have
nanoscale features and patterns, but are within a microscale form factor.
Furthermore, the biologically fabricated 63hotoluminescent nanostructured
silicon and germanium metal oxide materials have comparable optoelectronic
64
and chemical properties to industrially fabricated Si-Ge metal oxide materials.
These results show that highly 64hotoluminescent silica doped with Ge can be
fabricated biologically.
Acknowledgements
This research was supported by the National Science Foundation (NSF) award
number BES-0400648. We acknowledge Dr. Stephen Golledge of the
CAMCOR Surface Analytical Facility for XPS assistance.
Notes and references
a
School of Chemical, Biological, and Environmental Engineering, Oregon
State University, Corvallis, Oregon 97331 USA. Fax: 541-737-4600; Tel: 541737-3370; E-mail: galede@engr.orst.edu
b
Department of Physics, Portland State University, Portland, Oregon 97207
USA. Fax: 503-725-2815; Tel: 503-725-4228; E-mail: jiaoj@pdx.edu
1 L. Tsybeskov, J.V. Vandyshev, P.M. Fauchet, Phys. Rev. B: Condens.
Matter Mater. Phys. 1994, 4 , 7821-7824; b) M. R c schloss, T.
Wirschem, H. Tamura, G. Ruhl, J. Oswald, S. Vepře , J. Lumin. 1995, 63,
279-287; c) Y.D. Glinka, A.S. Zyubin, A.M. Mebel, S.H. Lin, L.P. Hwang,
Y.T. Chen, Chem. Phys. Lett. 2002, 358, 180-186; d) C.M. Carbonaro, F.
Clemente, R. Corpino, P.C. Ricci, A. Anedda, J. Phys. Chem. B 2005, 109,
14441-14444; e) D.J. Lockwood, J. Mater. Sci.: Mater. Electron. 2009, 20,
S235-S244.
2 a) T. Gao, X.M. Bao, F. Yan, S. Tong, Phys. Lett. A 1997, 232, 321-325; b)
J.K. Shen, X.L. Wu, X.M. Bao, R.K. Yuan, J.P. Zou, C. Tan, Phys. Lett. A
2000, 273, 208-211; c) J. Li, X.L. Wu, Y.M. Yang, X. Yang, X.M. Bao,
65
Phys. Lett. A 2003, 314, 299-303; d) A. Molle, M.N.K. Bhuiyan, G.
Tallarida, M. Fanciulli, Mater. Sci. Semicond. Process. 2006, 9, 673-678.
3 a) Y. Zhu, C.L. Yuan, S.L. Quek, S.S. Chan, P.P. Ong, Q.T. Li, J. Appl.
Phys. 2001, 90, 5318-5321; b) Y. Zhu, C.L. Yuan, P.P. Ong, J. Appl. Phys.
2003, 93, 6029-6033.
4 a) L. Rebohle, J. von Borany, R. Grӧtzschel, A. Mar witz, B. Schmidt, I. E.
Tyschen o, W. S orupa, H. Frӧb, K. Leo, Phys. Status Solidi A 1998, 165,
31-35; b) N. Arai, H. Tsuji, H. Nakatsuka, K. Kojima, K. Adachi, H.
Kotaki, T. Ishibashi, Y. Gotoh, J. Ishikawa, Mater. Sci. Eng., B 2008, 147,
230-234.
5 a) M. Zacharias, P.M. Fauchet, Appl. Phys. Lett. 1997, 71, 380-382; b)
X.M. Wu, M.J. Lu, W.G. Yao, Surf. Coat. Technol. 2002, 161, 92-95; c)
Z.W. Xu, A.H.W. Ngan, W.Y. Hua, X.K. Meng, Appl. Phys. A: Mater. Sci.
Process. 2005, 81, 459-463.
6 a) C.V. Cojocaru, F. Ratto, C. Harnagea, A. Pignolet, F. Rosei,
Microelectron. Eng. 2005, 80, 448-456; b) H.J. Fan, P. Werner, M.
Zacharias, Small 2006, 2, 700-717.
7 a) E. Dujardin, S. Mann, Adv. Mater. 2002, 14, 775-788; b) L.-Q. Wu, G.F.
Payne, Trends Biotechnol. 2004, 22, 593-599.
8 a) D. Losic, J.G. Mitchell, N.H. Voelcker, Adv. Mater. 2009, 21, 29472958; b) W. Yang, P.J. Lopez, G. Rosengarten, Analyst 2010, doi:
10.1039/c0an00602e.
9 a) W. Wang, T. Gutu, D.K. Gale, J. Jiao, G.L. Rorrer, C.-H. Chang, J. Am.
Chem. Soc. 2009, 131, 4178-4179; b) K.-C. Lin, V. Kunduru, M. Bothara,
K. Rege, S. Prasad, B.L. Ramakrishna, Biosens. Bioelectron. 2010, 25,
2336-2342; c) C. Jeffryes, R. Solanki, Y. Rangineni, W. Wang, C.-H.
Chang, G.L. Rorrer, Adv. Mater. 2008, 20, 2633-2637.
66
10 D.K. Gale, T. Gutu, J. Jiao, C.-H. Chang, G.L. Rorrer, Adv. Funct. Mater.
2009, 19, 926-933.
11 a) H.E. Townley, A.R. Parker, H. White-Cooper, Adv. Funct. Mater. 2008,
18, 369-374; b) L. De Stefano, A. Lamberti, L. Rotiroti, M. De Stefano,
Acta Biomater. 2008, 4, 126-130.
12 a) L. De Stefano, I. Rendina, M. De Stefano, A. Bismuto, P. Maddalena,
Appl. Phys. Lett. 2005, 87, 233902; b) A. Setaro, S. Lettieri, P. Maddalena,
L. De Stefano, Appl. Phys. Lett. 2007, 91, 051921; c) A. Bismuto, A.
Setaro, P. Maddalena, L. De Stefano, M. De Stefano, Sens. Actuators, B
2008, 130, 396-399; d) S. Lettieri, A. Setaro, L. De Stefano, M. De
Stefano, P. Maddalena, Adv. Funct. Mater. 2008, 18, 1257-1264; e) L. De
Stefano, L. Rotiroti, M. De Stefano, A. Lamberti, S. Lettieri, A. Setaro, P.
Maddalena, Biosens. Bioelectron. 2009, 24, 1580-1584.
13 a) M.R. Weatherspoon, S.M. Allan, E. Hunt, Y. Cai, K.H. Sandhage,
Chem. Commun. 2005, 651-653; b) S. Dudley, T. Kalem, M. Akinc, J. Am.
Ceram. Soc 2006, 89, 2434-2439.
14 a) D. Losic, G. Triani, P.J. Evans, A. Atanacio, J.G. Mitchell, N.H.
Voelcker, J. Mater. Chem. 2006, 16, 4029-4034; b) D.-H. Lee, T. Gutu, C.
Jeffryes, G.L. Rorrer, J. Jiao, C.-H. Chang, Electrochem. Solid-State Lett.
2007, 10, K13-K16.0.2; c) Y. Cai, M.B. Dickerson, M.S. Haluska, Z. Kang,
C.J. Summers, K.H. Sandhage, J. Am.Ceram. Soc. 2007, 90, 1304-1308; d)
D.-H. Lee, W. Wang, T. Gutu, C. Jeffryes, G.L. Rorrer, J. Jiao, C.-H.
Chang, J. Mater. Chem. 2008, 18, 3633-3635; e) T. Gutu, D.K. Gale, C.
Jeffryes, W. Wang, C.-H. Chang, G.L. Rorrer, J. Jiao, J. Nanomater. 2009,
860536; f) Y. Yang, J. Addai-Mensah, D. Losic, Langmuir 2010, 26,
14068-14072.
67
15 a) C. Jeffryes, T. Gutu, J. Jiao, G.L. Rorrer, Mater. Sci. Eng., C 2008, 28,
107-118; b) C. Jeffryes, T. Gutu, J. Jiao, G.L. Rorrer, ACS Nano 2008, 2,
2103-2112.
16 G. L. Rorrer, C.-H. Chang, S.-H. Liu, C. Jeffryes, J. Jiao, J.A. Hedberg, J.
Nanosci. Nanotechnol. 2005, 5, 41-49.
17 a) S. Oswald, B. Schmidt, K.-H. Heinig, Surf. Interface Anal. 2000, 29,
249-254; b) W.K. Choi, Y.W. Ho, S.P. Ng, V. Ng, J. Appl. Phys. 2001, 89,
2168-2172; c) B. Pelissier, H. Kambara, E. Godot, E. Veran, V. Loup, O.
Joubert, Microelectron. Eng. 2008, 85, 151-155.
18 T. Qin, T. Gutu, J. Jiao, C.-H. Chang, G.L. Rorrer, ACS Nano 2008, 2,
1296-1304.
19 T. Qin, T. Gutu, J. Jiao, C.-H. Chang, G.L. Rorrer, J. Nanosci. Nanotechnol.
2008, 8, 1-7.
20 a) H. Zhu, Y. Han, R.B. Wehrspohn, C. Godet, R. Etemadi, D. Ballutaud, J.
Appl. Phys. 1998, 83, 5386-5393; b) M.A. García, S. E. Paje, M.A.
Villegas, J. Llopis, Mater. Lett.2000, 43, 23-26; c) N. He, S. Ge, C. Yang,
C. Hu, M. Gu, Mater. Res.Bull. 2004, 39, 1931-1937; d) N.Y. He, S.X. Ge,
C. Yang, J.M. Cao, M. Gu, Mater. Lett. 2004, 58, 3304-3307.
21 a) H. Tamura, M. R c schloss, T. Wirschem, S. Vepře , Appl. Phys. Lett.
1994, 65, 1537-1539; b) L.T. Zhuravlev, Colloids Surf., A 2000, 173, 1-38;
c) Y. Inaki, H. Yoshida, T. Yoshida, T. Hattori, J. Phys. Chem. B 2002, 106,
9098-9106; d) L. Peng, W. Qisui, L. Xi, Z. Chaocan, Colloids Surf., A
2009, 334, 112-115.
22 P. Yuan, D.Q. Wu, H.P. He, Z.Y. Lin, Appl. Surf. Sci 2004, 227, 30-39.
23 N. Kröger, R. Deutzmann, M. Sumper, Science 1999, 286, 1129-1132.
24 a) V.B. Neustruev, J. Phys.: Condens. Matter, 1994, 6, 6901-6936; b) X.L.
Wu, T. Gao, X.M. Bao, F. Yan, S. S. Jiang, D. Feng, J. Appl. Phys. 1997,
82, 2704-2706; c) K. Prabhakaran, F. Maeda, Y. Watanabe, T. Ogina, Appl.
68
Phys. Lett. 2000, 76, 2244-2246; d) T.J. Grassman, S.R. Bishop, A.C.
Kummel, Surf. Sci. 2008, 602, 2373-2381.
25 J. Oh, J.C. Campbell, J. Electron. Mater. 2004, 33, 364-367.
69
Fig. 3-1
(a) Cell number density and soluble Si concentration in culture
medium as a function of time for the two-stage photobioreactor cultivation of
Pinnularia sp. (b) Soluble Si and Ge concentration in culture medium of Stage
II for an initial Ge concentration of 25 µm.
70
Fig.3-2
TEM images of Pinnularia sp. Obtained at the end of stage II
cultivation where no Ge was fed to the culture, the cells were silicon starved,
and the cell number density was constant for two photoperiods. (a) Intact
frustule; (b) submicron pore structure; (c) nanoscale pore array.
71
Fig. 3-3 TEM images and electron diffraction of Pinnularia sp. Obtained at
the end of the stage II cultivation which was doped with 0.49 weight % Ge and
annealed in air at 400ºC for 2 hours. (a),(b) Microstructure of Pinnularia sp.
Frustule; (c),(d) Patterned submicron pore array; (e),(f) and nanoscale pore
features with electron diffraction pattern inset.
72
Fig. 3-4.
Photoluminescence and thermal gravimetric analysis (TGA) as a
function of annealing temperature of (a) biosilica doped with 0.45 wt % Ge
and (b) native biosilica without Ge. (c) TGA profile of native biosilica weight
loss percent as a function of time for biosilica without Ge. Photoluminescence
signal normalized to the signal prior to annealing.
73
Fig. 3-5. (a) Normalized photoluminescence intensity and (b) peak
photoluminescence intensity wavelength of biosilica with 0, 0.24, 0.46, 0.49
and 0.96 wt % Ge in silica as a function of thermal annealing in air for 2 hours
at 250ºC, 400ºC, and 600ºC.
74
Fig. 3-6. Normalized photoluminescence spectrum of native biosilica without
Ge and biosilica doped with 0.96 weight percent Ge, which was annealed at
250ºC. The photoluminescence spectrum of the native biosilica was multiplied
by a factor of 5 for scale. Inset shown is a photograph of Pinnularia frustule
powder annealed at 250°C excited with 337 nm light.
75
Fig. 3-7. (a) X-ray photoelectron spectroscopy (XPS) analysis of industrial
GeO2 and Pinnularia sp. Biosilica doped with 0.96 wt % Ge which was as
deposited (no annealing), and annealed in air for 2 hours at 250ºC, 400ºC, and
600ºC. (b) XPS peak height intensity ratio of GeO to GeO2 before and after
annealing. The binding energy of GeO2 is 1220.2 eV and the binding energy of
GeO is 1221.6 eV.
76
Fig. 3-8. This wor was features on the front cover of the Journal of Materials
Chemistry, volume 21, number 2 , 2011.
.
77
Chapter 4
Germanium Enrichment in the Girdle Band of the Centric Diatom
Cyclotella sp.
Debra K. Gale*,†, Timothy Gutu‡, Jun Jiao‡, Chih-Hung Chang†, Gregory L.
Rorrer†
*,†
School of Chemical, Biological, and Environmental Engineering
Oregon State University, Corvallis, Oregon 97330 USA
galede@engr.orst.edu
‡
Timothy Gutu, Jun Jiao
Department of Physics, Portland State University
Portland, Oregon 97207 USA
78
ABSTRACT
There is significant interest in germanium (Ge) doped silica with nanoand micro-structured features for optoelectronic applications. In this study, Ge
was metabolically enriched in the girdle band biosilica of the diatom Cyclotella
sp. to yield a frustule, which from the valve face view appeared as a patterned
silica disc, encircled with a sub-micron in thickness Ge ring. Diatoms are
single celled photosynthetic algae that make silica shells called frustules. In
this study, a two stage photobioreactor cultivation strategy was used to
metabolically enrich Ge in the girdle band of the frustule. In stage I, the
diatom cells were grown to silicon (Si) starvation and stationary phase. In
stage II, the diatom cells were fed a mixture of Si and Ge for 48 hours, or 2
photoperiods. ICP and STEM-EDS confirmed that Ge was doped into the
frustule biosilica up to a level of 0.35 weight percent. Mapping by microRaman spectroscopy and mathematical prediction showed that Ge was
spatially enriched in the girdle band of the Cyclotella frustule by a factor of 3
times compared to the Ge in the valve. This is the first study to use a strategic
cultivation strategy to metabolically enrich a foreign dopant into a specific
spatial location in a diatom frustule. This study provides evidence that three
dimensional Si microstructures spatially doped with Ge, which have nano and
submicron features, can be fabricated biologically.
KEYWORDS
Diatom; Enriched; Germanium; Photobioreactor; Raman;
79
INTRODUCTION
There has been significant interested in the fabrication of germanium
(Ge) doped silica with spatially ordered features, which exhibit enhanced
optoelectronic properties with application to the next generation of display
devices, semiconductors, and sensors. [1-3] Conventional top down methods
of fabrication such as ion implantation, [4, 5] co-sputtering, [6] and laser
ablation [7] can produce planar silica doped with targeted concentrations of
Ge, which exhibit interesting optical properties. Since the discovery that
planar silica doped with Ge exhibits unique properties, the interest in 3dimensional structures with nano- and microscale features has been piqued. [810] However, the fabrication of 3-dimensional materials using 1-dimensional
nanostructures as building blocks is challenged due to the requirement for
advanced fabrication techniques, and the lack of control over chemical
composition and morphology. [8, 9, 11, 12]
To address the limitations of traditional 3-dimensional top down
semiconductor fabrication techniques, interest in bio inspired hierarchical
material synthesis has emerged. [13] Specifically, diatoms, single celled
photosynthetic algae, fabricate nano- and micro-structured hierarchical silica
shells, called frustules, from bottom up synthesis. These frustules are
comprised of nanoscale patterned pore arrays within a microscale form factor
and also exhibit unique optoelectronic properties, [14,15] features desirable for
a three dimensional semiconductor material. Diatoms are emerging as a lowcost, high throughput method to fabricate nanostructured 3-dimensional silica
materials which have potential to be integrated into future micro-scale devices.
[16-20]
In this study, we demonstrate that Ge can be preferentially doped into
the girdle band of the centric diatom Cyclotella sp. using a two-stage
photobioreactor cultivation strategy. Recently we discovered that Ge could be
80
metabolically doped uniformly throughout the newly formed frustule biosilica.
[21, 22] However, this study is unique because it is the first to show how the
diatom cell can be manipulated to insert Ge into a specific submicron spatial
location in the frustule. From the valve face perspective, the frustule appears
as a silica disk, which contains a patterned porous array, encircled with a Ge
ring, submicron in thickness. This is the first example to show that a living
organism can be used to fabricate nano- and micro-structured silica with
spatially enriched Ge.
EXPERIMENTAL SECTION
Diatom Cell Culture. Axenic cultures of the photosynthetic marine diatom
Cyclotella sp. were obtained from UTEX The Culture Collection of Algae
(#126 ). Maintenance culture of this organism in Harrison’s Artificial
Seawater Medium [23] was previously described. [15]
Photobioreactor Cultivation. The 6.1 L total volume bubble column
photobioreactor for cultivation of Cyclotella sp. cells was previously described.
[21] The photobioreactor was modified with a two-channel syringe pump
(World Precision Instruments, Aladdin 8000) for continuous delivery of a
mixture of dissolved silicon and germanium. Cultivations were carried out at
150 µE m-2 s-1 incident light intensity, 14 hour light/10 hour dark photoperiod,
0.10 L air min-1 aeration rate (~350 ppm CO2), and a constant temperature of
22°C.
The bioreactor cultivation process was carried out in two stages. In
Stage I of the cultivation process, 4.5 L of Harrison’s Artificial Seawater
Medium was inoculated to 2.0x105 cells mL-1 from flask cultured Cyclotella
sp. diatoms which were harvested in the stationary phase of growth. The
81
photobioreactor medium contained 0.65 mM soluble silicon (Si) from
Na2SiO3, which targeted three cell doublings in Stage I. The pH of the culture
medium began at approximately 8.43 and at the end of Stage I had increased to
approximately 8.9. The end of Stage I was identified by silicon starvation,
which was defined when all of the soluble silicon was consumed and the cell
number density was constant for at least two photoperiods. In Stage IIA of the
cultivation process, a soluble mixture of 31.25 mM Si and 0.47 mM Ge in
distilled/deionized water, from Na2SiO4 and GeO2, was delivered into the
photobioreactor medium, controlled by the syringe pump, at a volumetric flow
rate of 3 mL hr-1 for 48 hours, which was 2 photoperiods beginning at 9:00
AM when the light phase began. The amount of silicon fed to the
photobioreactor targeted one cell doubling in Stage II. Stage IIB was defined
as the time after the 48 hour Si and Ge addition until the end of the
photobioreactor cultivation experiment. The pH in Stage II ranged from 9 to
9.6 pH units. The control cultivation experiment was executed in two Stages,
like what was previously described, except there was no GeO2 added in Stage
II.
The time course of the Cyclotella sp. two stage photobioreactor
cultivation was monitored by measurements of cell number density in triplicate
with a Beckman Z2 Coulter Counter fit with a 100 µm aperture, and
measurements of Si and Ge in the culture medium, which was measured
spectrophotometrically as described previously. [21] In Stage I of the
cultivation, a 40 mL cell volume was removed from the bioreactor every 24
hours for measurement of cell number density and silicon concentration,
whereas in Stage II sampling was increased to every 8 hours and Ge was also
measured. A cell culture volume of 700 mL was removed at 0, 24, 38, 48, 72,
and 120 hours into Stage II to determine the silicon and germanium content in
the dry cell biomass and frustule biosilica.
82
Frustule Biosilica Isolation and Ge Analysis. Biosilica frustule isolation
from the Cyclotella sp. cells was previously described. [21, 15] The diatom
cells were treated with 30 weight % aqueous hydrogen peroxide at pH 2.5 to
remove the organic materials and isolate the intact biosilica frustule. One gram
of dried cell mass typically yielded 0.10 to 0.13 grams of frustule biosilica.
The silicon and germanium content in the solid dry cell mass and biosilica was
measured by Inductively Coupled Plasma (ICP) following NaOH fusion of the
solid as previously described. [21]
Electron Microscopy. Diatom frustules were imaged by transmission
electron microscopy (TEM) and scanning transmission electron microscopy Xray dispersion spectroscopy (STEM-EDS). Biosilica frustules isolated from
diatom cells were suspended in methanol, then deposited onto a holey carbon
coated copper TEM grid and allowed to air dry. Frustules were then imaged
by TEM or STEM-EDX at 200 keV with an FEI Tecnai F20 TEM.
Raman Spectroscopy. Raman maps and line scans of the Cyclotella sp.
frustules were acquired with a Horiba Jobin Yvon Lab Ram HR Confocal
Raman Microscope. Frustules were deposited on a calcium fluoride window
(Edmund Industrial Optics, #NT48-854, 15 mm diameter) which exhibited a
low Raman background. Intact individual frustules with attached girdle bands
were excited with a 50 mW, 532 nm incident laser excitation source through a
100x (0.9 N.A.) objective of the Olympus BS41 microscope, which was
manually focused to yield the highest Raman signal intensity for collection.
The confocal pinhole was set to a diameter of 150 µm which minimized
background signal and maximized biosilica Raman signal collection by a CCD
with a 1200 lines mm-1 grating. Raman line scans and maps were collected
83
using the Horiba DUOScan scanning module, with a 0.350 µm step size, 120
second accumulation time, and collected in the Raman spectral range from
300cm-1 to 1100 cm-1. Each point of the Raman line scan was comprised of 2
averaged Raman spectra, where as each Raman mapping point was 1 Raman
spectra. Maps were acquired with a 26 x 29 point mapping array. One
Cyclotella sp. frustule fit within the maximum mapping area limited by the
objective. Spectra were recorded and processed with Horiba LabSpec 5 data
acquisition and analysis software, which was used for baseline subtraction and
peak identification. At least 5 Raman line scans and 2 maps were collected for
each sample.
GE ENRICHMENT PREDCITION
A material balance model was developed to calculate the amount of
enriched GeO2 in the daughter girdle band (hypotheca) compared to the
daughter valve (hypovalve). Terminology for components of the diatom
frustule are presented in Figure 7a. It was not possible to distinguish between
the daughter and parent components of the frustule in Figure 7a, so the
daughter girdle bands and valve were assigned. Governing mass balance
equations were developed for the Si and Ge in the biosilica at the end of stages
IIA and IIB. The mass balances were developed using some simplifying
assumptions of the bioreactor and of the cell cycle phases in stages IIA and
IIB, which are shown as a schematic in Figure 3a. It was assumed that the
population of Si starved cells in the well mixed photobioreactor arrested in the
late G1 phase of cell division at the end of Stage I. [24, 25] In the late G1
phase of cell division, the entire daughter epitheca has been deposited and any
Si or Ge uptake after this point is directed towards fabrication of the newly
formed frustule hypotheca. [25] It was assumed that the G2 and M phases of
the cell cycle were completed during Stage IIA. During the G2 and M cell
84
cycle phases, the hypovalve has been deposited and exocytosed from the
silicon deposition vesicle and fabrication of the girdle band has been initiated.
[25] The newly formed diatom cells will not separate unless the hypovalves
have been formed. [26, 27] Since the diatoms cells separated in stage IIA, it
was assumed that the G2 and M phases of the cell cycle were also completed in
stage IIA. The final phase of the cell cycle, G1, was assumed to take place in
stage IIB of the photobioreactor cultivation such that all Si and Ge deposition
into the frustule only went towards finishing the fabrication of the daughter
girdle band (hypocingulum). Finally, all Si and Ge in the biosilica was
assumed to be in the form of GeO2 and SiO2.
The volume of the daughter girdle bands, VGB (hypocingulum),
daughter valve, VV (hypovalve), and frustule presented in Equations (3), (4),
and (5) were calculated from measurements estimated from the SEM image in
Figure 7a, and are shown in Figure 7b. The pore volume was not accounted
for in the frustule volume calculation. A governing equation for the ratio of
GeO2:SiO2 in the girdle band compared to the valve is presented in Equation
(6), developed from equations (1)-(5).
∑
∑
(1)
∑
∑
85
(2)
( )
( )
(
)
(3)
(4)
(5)
(
)
(
)
(6)
Nomenclature
CGe,DW,t
CGe,m,t
CGe,r
CSi,DW,t
CSi,m,t
CSi,r
dGB
dV
hGB
hV
MGe,BS,IIA
MGe,BS,IIB
mGe,cell,IIA
Ge concentration in the dry cell weight at time t (mg Ge/ mg
DW)
Ge concentration in the bioreactor medium at time t (mmol
Ge/L)
concentration of Ge in the perfusion reservoir (mmol Ge/L)
Si concentration in the dry cell weight at time t (mg Si/ mg DW)
Si concentration in the bioreactor medium at time t (mmol Si/L)
concentration of Si in the perfusion reservoir (mmol Si/L)
girdle band diameter (µm)
valve diameter (µm)
height of the girdle bands, epicingulum and hypocingulum (µm)
height of the valve (µm)
mass of Ge in the biosilica at the end of stage IIA (mg Ge)
mass of Ge in the biosilica at the end of stage IIB (mg Ge)
mass of Ge in the cell, not incorporated into the biosilica at the
end of stage IIA (mg Ge)
86
mGe,cell,IIB
mGe,M,IIA
mGe,M,IIB
MGe/Si, IIA
MGe/Si, IIB
mSi,cell,IIA
mSi,M,IIA
mSi,M,IIB
mSi,cell,IIB
MWGe
MWSi
MSi,BS,IIA
MSi,BS,IIB
RGB/Valve
t
tf
tGe
tSi
VFrust
VGB
vGe
VM,s,t
vSi
VV
XDW,t
mass of Ge in the cell, not incorporated into the biosilica at the
end of stage IIB (mg Ge)
mass of Ge in the bioreactor medium at the end of stage IIA
(mg Ge)
mass of Ge in the bioreactor medium at the end of stage IIB (mg
Ge)
mass ratio of Ge to Si in the biosilica at the end of stage IIA
(mg Ge/mg Si)
mass ratio of Ge to Si in the biosilica at the end of stage IIB (mg
Ge/mg Si)
mass of Si in the cell, not incorporated into the biosilica at the
end of stage IIA (mg Si)
mass of Si in the bioreactor medium at the end of stage IIA (mg
Si)
mass of Si in the bioreactor medium at the end of stage IIB (mg
Si)
mass of Si in the cell, not incorporated into the biosilica at the
end of stage IIB (mg Si)
molecular weight of Ge (g/mol)
molecular weight of Si (g/mol)
mass of Si in the biosilica at the end of stage IIA (mg Si)
mass of Si in the biosilica at the end of stage IIB (mg Si)
mass ratio Ge:Si in the biosilica of the girdle band
(hypocingulum) compared to the valve (hypovalve) at the end
of Stage IIB
time reference of bioreactor sampling
frustule wall thickness (µm)
time of Ge addition (hrs)
time of Si addition (hrs)
volume of frustule (µm3)
volume of daughter girdle bands, hypocingulum (µm3)
volumetric flowrate of Ge (mmol Ge/L)
volume of medium sampled at time t (L)
volumetric flowrate of Si (mmol Si/L)
volume of daughter valve, hypovalve (µm3)
dry cell weight density in the bioreactor at time t (g/L)
87
RESULTS
Localized Ge enrichment in the diatom girdle band
Ge was metabolically inserted into the girdle band of the Cyclotella sp.
diatom biosilica using a two stage cultivation strategy. The photobioreactor,
which we described previously, [21, 28] consisted of a bubble column vessel to
mix and aerate the cell suspension, an external light stage, and a syringe pump
for the controlled delivery of a mixture of soluble Si and Ge.
In Stage I of the cultivation process, the photobioreactor which
contained an initial cell density of 2x105 cells mL-1 was charged with 0.65 mM
soluble Si, to allow for approximately 3 cell doublings. A summary of
cultivation parameters for the control cultivation where no Ge was added and
for the cultivation where Ge was added in Stage II is presented in Table 1.
The end of Stage I of the two stage cultivation process was defined by silicon
depletion in the photobioreactor medium and a constant cell number density
for at least 2 photoperiods, 48 hours.
In Stage IIA of the cultivation, a solution of soluble Si (31.25 mM) and
Ge (0.47 mM) was fed by a syringe pump to the silicon starved cells at a rate
of 3 mL hr-1 for 2 photoperiods (48 hours) or 24 hours, at the start of the light
cycle in the photoperiod. Stage IIB began when the addition of Si and Ge to
the photobioreactor was stopped.
The Si addition targeted 1 cell doubling in
Stage IIA. Si was used to control the cell growth in both stages because it is a
required substrate for diatom cell division. [21, 22, 28] The cell number
density and Si concentration for a representative experiment which was
delivered Ge in stage II for 48 hours and 24 hours is presented in Figure 1a
and Figure 1b. Dotted lines in these figures indicate the start and stop of the
continual addition of Si and Ge in Stage IIA. The concentration profile of Si
and Ge in the photobioreactor medium in stage II, which corresponds to the
88
cultivations in Figures 1a and 1b, is presented in Figures 2a and 2b. The Si
and Ge concentration profile in the bioreactor medium if there was no uptake
by the cells is also presented in Figures 2a and 2b.
The diatom cells doubled in concentration within the 48 hour stage IIA
(Figure 1a) , but not within the 24 hour stage IIA (Figure 1b). Although there
was uptake of Si and Ge by the cells in the 48 hour stage IIA, the residual Si
and Ge in the bioreactor medium increased through the 48 hour addition. A
control experiment, where no Ge was added in stage II is presented in Figures
1c and 2c. The cell density increased after Si and Ge addition was stopped,
unlike the cultivation with Ge in Figures 1a and 1b. Following cell division
and Si/Ge addition in Figures 1a and 1b, the cells continued to take up Si and
Ge from the photobioreactor medium as shown by the concentration profiles in
Figures 2a and 2b. Ge uptake and metabolism by the diatom cells did not
significantly affect the growth cultivation parameters as shown in Table 1,
except for a slight decrease in the cell yield in stage II, which means that the
amount of Si per cell increased.
Germanium doped diatom biosilica
Throughout the duration of the 48 hour stage IIA, diatom cells were
removed from the photobioreactor for analysis of the Ge content in the
biosilica. The Ge content was analyzed prior to Si/Ge addition in stage II, and
24, 38, 48, 72, and 120 hours after the commencement of 48 hour Si/Ge
addition. Biosilica frustules were extracted from the diatom cells by treatment
with aqueous hydrogen peroxide at a pH of 2.5 and measured for the Ge
content in the biosilica by ICP, as described in the experimental section. The
weight percent of Ge measured in the frustule biosilica during stage II is
presented in Figure 3a. Commensurate with a decrease of the Si/Ge
concentration in the photobioreactor medium in Figure 2a, the Ge
89
concentration in the biosilica increased throughout the duration of Stage II to
yield final Ge concentration of 0.359 ± 0.027 weight percent. The predicted
mass ration of Ge to Si in the frustule daughter girdle band and valve from
equations (1) – (6) is presented in Figure 3b.
STEM-EDS was performed as further confirmation that Ge had been
metabolically doped into the Cyclotella frustule. A representative STEM-EDS
spot analysis of a frustule from 120 hours into stage IIA, is presented in Figure
4. At least three spots on five frustules were analyzed with EDS spot scans for
the presence of Ge which has a characteristic Kα energy of .86 eV. The spot
denoted with a 1 in the center of the valve in Figure 4a is the location of the
EDS spot scan for the data shown in Figure 4b. STEM-EDS line scans were
not used to characterize the enriched Ge in the girdle band because the
biosilica was thickened near the perimeter of the valve facing frustule, which
blocked electron transmission.
Electron microscopy of diatom frustules
Cyclotella frustules isolated from cells collected at the end of stage I,
which did not contain Ge, were imaged by TEM as shown in Figure 5.
Cyclotella sp. is a centric diatom which has two disc shaped valves, the
hypovalve and epivalve, that are connected with ring shaped girdle bands, the
hypocingulum and epicingulum. The frustule is approximately 10 µm in
diameter and 6 µm thick. A TEM image of a Cyclotella frustule, valve side
facing, is presented in Figure 5a. The frustule valve is made up of radially
symmetrical patterned pores. Pores which are approximately 100 nm in
diameter extend from the center of the frustule to the valve perimeter, where
they are met with larger 200 nm in diameter pores, called the rimoportulae.
Lining the base of each 100 nm pore are four to five 20 nm in diameter pores.
A TEM image of the pore detail located near the center of the valve is shown
90
in Figure 5b. Off center of the middle of the valve is one large pore, the
fultoportulae, which has approximately the same diameter as the rimoportulae.
The hydrogen peroxide treatment method used to oxidize cell organics
sometimes separated the girdle band from the valve. The valves separated
from the girdle band were chosen for TEM imaging. Frustule valves that did
not separate from the girdle band were difficult to image due to the frustule
thickness, which interfered with electron transmission.
TEM images of representative frustules extracted from cells at 120
hours into stage II of the cultivation which was fed Ge for 48 hours, is
presented in Figure 6. Upon cell division, the diatom cell fabricates two new
daughter valves. As a result, the new daughter cells have one parent valve
(epivalve) made up of silica, and one daughter valve (hypovalve) which
contains metabolically doped Ge. Statistically, only half of the valves contain
metabolically inserted Ge. The bulk Ge concentration measured in the frustule
biosilica shown in Figure 5 was 0.359±0.027 weight percent, which
represented an average between the parent valves without Ge and the daughter
valves which contained Ge. Metabolic insertion of Ge into the biosilica of the
diatom cell altered the frustule fine features. The morphology of the
submicron pores which extended from the valve center to the valve perimeter
were altered in size and in the radial pattern on the valve surface. The 100 nm
pores which extend from the valve perimeter to the valve center appeared to
decrease in diameter and eventually close, in proximity to the valve center.
Furthermore, the four to five 20 nm pores which were set within the 100 nm
pores were not present in the frustule doped with Ge. The radial pore pattern
on the valve surface, Figure 6a and 6b, appeared aberrant compared to the
pore pattern on the valve surface without Ge, Figure 5. Instead of a radially
symmetrical patterned pores present in the valves without Ge, the pores on the
91
valves doped with Ge exhibited pore arrays which extended almost
perpendicular to the radial pores.
Micro Raman analysis of Ge enrichment in the girdle band
The Ge enrichment in the daughter girdle band of the cyclotella frustule
was estimated using the mass balance model and measured with Raman
microscopy. The predicted ratio of Ge:Si enriched in the daughter girdle band
of the biosilica estimated from equation (4) was 3.2. These modeling results
suggested that there was 3.2 times more Ge in the daughter girdle band than in
the valve. The measured ratio of Ge:Si in the girdle band compared to the
valve was 3.63 ± 1.07 measured from 9 Raman line scans.
The Ge distribution throughout the Cyclotella frustule was
characterized with Raman Microscopy analysis line scans and mapping.
Diatoms chosen for analysis by Raman spectroscopy were the hypotheca, with
observable attached and intact girdle bands, which were extracted from diatom
cells at 120 hours into stage II of the cultivation fed with Ge for 48 hours.
Frustules that did not contain Ge were extracted from cells at the end of stage
II in the control cultivation. Frustules were deposited on a calcium fluoride
(CaF) window for Raman analysis. CaF has a low spectral background except
for an intense peak at 322 cm-1. Frustules were excited with a 532 nm, 50 mW
excitation source which was manually focused on the frustule surface.
LabSpec software provided by Horiba Jobin Yvon was used to collect Raman
spectra and process the spectra by linear baseline subtraction, peak detection,
and peak fitting to obtain the peak amplitude. The only post-processing on the
Raman data was linear baseline subtraction.
A representative Raman spectrum extracted from a line scan of a
frustule doped with Ge is presented in Figure 8a. The peaks at 490 cm-1, 800
cm-1, and 970 cm-1 are related to vibrational states of silica, which has been
92
shown for industrial silica (SiO2), [29, 30, 32] and diatom biosilica. [31, 33]
The portion of the spectra which shows the germanium dioxide (GeO2) Raman
peak of a frustule doped with Ge is shown in Figure 8b. The peak at 420 cm-1,
which is on the shoulder of the 490 cm-1 peak shown in Figure 8b is assigned
to the symmetric stretching of germanium dioxide. [34-37] Peaks were fit
using the LabSpec 5 Raman software with Gaussian peaks. The Raman
spectrum of a frustule without Ge from the control cultivation is shown in
Figure 8c. There is no Ge peak in the spectrum of the frustule without Ge.
Figures 8b and 8c were normalized to 1.0 to show relative peak size and
intensity because the signal counts were not always the same for every frustule.
Raman line scans were acquired to map the Ge distribution in the
diatom frustule valve and girdle band from the cultivation with a 48 hour stage
IIA. Line scans which show the relative Raman peak intensity of Ge to Si (420
cm-1/ 970 cm-1) of frustules from the cultivation fed with Ge which were doped
with Ge (Figure 9a, 9b), from the cultivation fed with Ge where the parent
valve and girdle band did not contain Ge (Figure 9c), and from the control
cultivation which was not fed Ge (Figure 9d) are all presented in Figure 9.
Line scans were acquired from valve facing frustules which had visible
attached girdle bands as shown in the inset of Figure 9c. The Ge signal
intensity was normalized to the silica Raman signal intensity to account for
varying amounts of material throughout the valve and girdle band. The Ge
signal was the most intense at the location of the girdle bands as shown in
Figures 9a and 9b. Although a low level of Ge was present throughout the
valve, also shown by STEM-EDS in Figure 4, the Raman line scan of the
whole frustule confirmed that the Ge concentration was enriched in the girdle
band of the frustule.
Raman maps were used to qualitatively observe the Ge and Si location
throughout the diatom frustule. A representative Raman map of a Cyclotella
93
frustule is presented in Figure 10. The integrated signal intensity of Raman
signals associated with silica 970 cm-1 and 490 cm-1, is shown in blue and red
respectively, in Figures 10a and 10b. The integrated Raman signal intensity
associated with Ge is shown in green in Figure 10c. The lateral resolution was
estimated to be 0.7 µm and the sample depth analyzed was approximately 3-4
µm. The CaF substrate was detected below the frustule for every
measurement, even below the thickest girdle band portion of the frustule,
which led us to assume that the entire frustule epitheca thickness was analyzed
for Raman line scans and mapping. Although the lateral resolution limited the
ability to detect fine structural details shown in Figures 5 and 6, submicron
structural details of the girdle band were detected by Raman mapping, as
shown in Figure 10 by the scalloped perimeter of the frustule.
DISCUSSION
This study has shown that 3-dimensional, nano- and micro-structured
silica doped with spatially localized Ge, can be fabricated by a biological
platform of living diatom cells. This is not the first study to intentionally dope
diatom frustule biosilica with a foreign metal or semiconductor such as Ge,
Titanium, or Nickel to alter the overall nano- and micro-structure and elicit
interesting optical properties. [21, 22, 28, 38, 39] However, this is the first
study to externally control the spatial location of Ge deposition by the diatom
cell into the girdle band of the frustule.
Ge was targeted into the girdle band of the Cyclotella diatom frustule
by a two stage photobioreactor cultivation strategy which co-fed a mixture of
Si and Ge to the diatom cells over a time of two photoperiods to yield one cell
division. Upon frustule (valve and girdle band) fabrication, the diatom cell
selectively enriched the daughter girdle band (hypocingulum) of the frustule
94
with Ge. Enrichment of the girdle band with Ge produced a diatom frustule
which from the valve face view, appeared as a nano-patterned silica disc,
encircled with a ring of Ge, sub-micron in thickness.
The photobioreactor cultivation strategy was developed to exploit the
tight coupling between cell division and frustule deposition by the cell. [27,
40] In stage I of the cultivation strategy cells were fed enough silica to allow
for 3 cell doublings as shown in Figure 1. The completion of stage I was
defined by silicon depletion in the photobioreactor medium and stationary
phase for at least 2 photoperiods. Si starvation of the centric diatoms
Thalassiosira weissflogii [25, 41] and Thalassiosira pseudonana, [24] which
are in the same classification order Thalassiosirales as cyclotella sp., caused
arrest in the G1 phase of the cell cycle. [42, 43] During the G1 phase of the
diatom cell cycle the cell has completed cell division, but the cell is still
growing and silica is deposited into the girdle bands. [25] Upon Si and Ge
replenishment to the starved culture in stage IIA, the cells pass through the S
phase where DNA replication occurs and enter the G2 and M phases of cell
division which are defined by hypovalve synthesis and exocytosis from the
silicon deposition vesicle and the initiation of girdle band fabrication. [24]
Immediately following Si uptake in stage IIA by the cell in the form of
Si(OH)4 it is condensed and transported to the silica deposition vesicle (SDV),
where silica nanoparticles are formed and patterned into the nano-porous
ornately patterened frustule. [44, 26] Centric diatoms deposit the frustule
valve in the Ge + M phase of the cell cycle (stage IIA) and then the girdle band
in the G1 phase of the cell cycle (stage IIB) in sequential order in a separate
SDV. [45, 24]
Deposition of the new daughter valve (hypovalve) is initiated
in the center of the valve face and is deposited radially. [45] Once the
hypovalve has been deposited, the SDV for the girdle band is formed. The
daughter cells are mature enough to separate before the girdle band has been
95
deposited. Following cell separation the girdle band will be formed in the G1
phase of the cell cycle, which we targeted to occur during stage IIB of the cell
cycle.
Si and Ge were added for a total of 2 photoperiods in stage IIA for 2
reasons. First, the 2 photoperiod Si and Ge addition maintained the diatom
culture in a state close to Si starvation which imparted external control of
Si/Ge uptake by the cell and deposition into the frustule, such that they likely
occurred almost simultaneously. [40] The two photoperiod Si and Ge addition
was commensurate with the average time of one Cyclotella cell doubling, 3540 hours. Previously we showed that when the pennate diatoms Pinnularia sp.
and Nitzschia frustulum are in a Si starved state and charged with a mixture of
Si and Ge, the cells will use a surge uptake mechanism of the Si and Ge into
the cell within hours of delivery. [21, 22] This type of substrate surge uptake
mechanism deposited Ge uniformly throughout the daughter valve and girdle
band of the frustule, and this deposition was ultimately only controlled by the
diatom cell. Second, we hypothesized that the Si and Ge in the bioreactor
medium in stage IIB, after cell division, would be deposited into the daughter
girdle band. This is evidenced by the division of the Cyclotella cells in stage
IIA and the continual uptake of Ge in stage IIB which is shown in the Figure
7c schematic. The two photoperiod Si and Ge addition was targeted to sustain
the cells in a starved state, even though the concentration of Si and Ge in the
photobioreactor medium increased throughout the addition phase in stage IIA.
The Si and Ge increased the most during the dark phases of the photoperiod
which occurred between 14-24 and 36-48 hours in stage IIA. Details of silicon
uptake of Cyclotella throughout the light and dark phase of the photoperiod
have not been documented in the literature. The coupling between Si/Ge
uptake and the photoperiod was observed in our previously work, [21, 22] and
occurs because in many species silicon metabolism has been shown to be
96
linked with photosynthetic activity which occurs during the light phase of the
photoperiod. [40] The two photoperiod addition of Si and Ge targeted one cell
division during stage IIA as shown in Figure 1a. To confirm that a 48 hour
stage IIA window was optimal, Si and Ge were added to the photobioreactor
for one photoperiod as shown in Figure 1b. However, one cell division was
not completed during the shorter stage IIA. The two photoperiod stage IIA
harnessed the cell growth cycle shown in the Figure 3a schematic. Cell
division and valve fabrication in stage IIA directed Ge deposition into the
girdle band of the frustule in stage IIB.
The STEM-EDS point analysis in Figure 4, the observation of abberent
pore morphology in Figure 6, and Raman mapping in Figures 9 and 10
confirmed that low levels of Ge were deposited into the valve of the diatom
frustule, but selectively enriched in the girdle band.
As targeted, the diatom selectively enriched Ge in the girdle band of the
frustule, as evidenced by the Raman line scans in Figure 9. In our previous
studies when Ge was metabolically inserted into the frustule using a surge
uptake mechanism of Si and Ge delivery, the Ge was uniformly deposited
throughout the entire daughter valve and girdle band. [21, 22] However, when
Si and Ti were co-fed to the diatom culture Pinnularia sp. over a period of 10
hours, the Ti was selectively enriched in the nanoscale fine structures located
in the frustule pores. [28] Like the girdle band in the centric diatom
Cyclotella, the nanoscale fine structures in the Pinnularia frustule are also
associated with the final stages of Si deposition. It was beyond the scope of
this study to do a parametric study of Si and Ge deposition under externally
controlled conditions. However, this study is demonstration that a cultivation
strategy can be designed with the cell cycle to control more than one substrate
component in frustule fabrication.
97
Representative Raman line scans of diatom biosilica doped with Ge
presented in Figure 9a and 9b, show the inherent variability of Si and Ge
deposition in the diatom frustules. The common feature between these line
scans is the enrichment of Ge in the daughter girdle band by approximately a
factor of 3 compared to the hypovalve of the frustule. This Raman data is
agreed with the Ge enrichment calculated from Equation (6) which was
derived from a material balance and frustule volume measurement. Like the
centric diatom Thalassiosira weissfloggi, [25] we estimated that approximately
40% of cellular silica is contained within the girdle bands, and 60% is
contained within the valves. Quantitative analysis of the absolute amount of
Ge in the diatom frustule by Raman spectroscopy is inherently complicated
due to the difficulty in determining sample specific parameters such as
scattering volume and Raman cross section. [46] The Raman cross section is
measurement of the Raman emission efficiency for each chemical species.
The relative Raman cross section of pure GeO2 has been reported to be on the
order of 9 times that of SiO2. [35] Although GeO2 is a more intense Raman
scatterer, this may have been to our advantage because the Ge concentration in
the biosilica was inherently low. Despite the inherent difficulty in Raman
quantification, there is precedence for qualitatively comparing the composition
of two species by normalizing the Raman peak height intensity of Raman
bands, [47, 48] as we did in this study. The Raman line scans, presented in
Figure 9, confirm that the composition of the daughter girdle is enriched with
Ge.
CONCLUSIONS
In this study, we demonstrated that diatom cells can be used as a platform to
fabricate ornately patterned silica microstructures doped with Ge spatially
98
enriched in the girdle band of the frustule. A two stage photobioreactor
cultivation strategy was developed which complimented the cell cycle to
enrich Ge in the daughter girdle band by a factor of 3 compared to the valve of
the diatom frustule as shown by Raman microscopy. This experimental result
agreed with the predicted enrichment obtained by a material balance model of
the Si and Ge. From the valve face view, the frustule looked like a patterned
silica disc, encircled with a sub-micron Ge ring. This is the first study to
metabolically insert enriched Ge into a specific spatial location in the diatom
frustule. This study provides evidence that three dimensional Si
microstructures spatially doped with Ge, which have unique nano and
submicron features, can be fabricated biologically.
ACKNOWLEDGEMENTS
This research was supported by the National Science Foundation (NSF) award
number BES-0400648.
REFERENCES
[1]
Alessi, A.; Agnello, S.; Ouerdane, Y.; Gelardi, F.M. J. Phys.-Condens.
Mat. 2011, 23, 1-6.
[2]
Grandi, S.; Mustarelli, P.; Agnello, S.; Cannas, M.; Cannizzo, A. J. SolGel Sci. Techn. 2003, 26, 915-918.
[3]
Huang, J.-G.; Lee, C.-L.; Lin, H.-M.; Chuang, T.-L.; Wang, W.-S.;
Juang, R.-H.; Wang, C.-H.; Lee, C.-K.; Lin, S.-M.; Lin, C.-W. Biosens.
Bioelectron. 2006, 22, 519-525.
[4]
Gao, T.; Bao, X.M.; Yan, F.; Tong, S. Phys. Lett. A. 1997, 232, 321325.
99
[5]
Zatsepin, A.F.; Fitting, H.-J.; Kortov, V.S.; Pustovarov, V.A.; Schmidt,
B.; Buntov, E.A. J. Non-Cryst. Solids 2009, 355, 61-67.
[6]
Shen, J.K.; Wu, X.L.; Bao, X.M.; Yuan, R.K.; Zou, J.P.; Tan, C. Phys.
Lett. A. 2000, 273, 208-211.
[7]
Zhu, Y.; Yuan, C.L.; Quek, S.L.; Chan, S.S.; Ong, P.P. J. Appl. Phys.
2001, 90, 5318-5321.
[8]
Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.;
Kim, F.; Yan, H. Adv. Mater. 2003, 15, 353-389.
[9]
Hu, J.; Jiang, Y.; Meng, X.; Lee, C.-S.; Lee, S.-T. Small, 2005, 1, 429438.
[10]
Costacurta, S.; Malfatti, L.; Kidchob,T.; Takahashi, M.; Mattei, G.;
Bello,V.; Maurizio, C.; Innocenzi, P. Chem. Mater. 2008, 20, 32593265.
[11]
Gu, Z.; Liu, F.; Howe, J.Y.; Paranthaman, M.P.; Pan, Z. Crys. Growth
Des.2009, 9, 35-39.
[12]
Barth, S.; Hernandez-Ramirez, F.; Holmes, J.D.; Romano-Rodriguez,
A. Prog. Mater. Sci. 2010, 55, 563-627.
[13]
Dujardin, E.; Mann, S. Adv. Mater. 2002, 14, 1-14.
[14]
Qin, T.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. J. Nanosci.
Nanotechno. 2008, 8, 1-7.
[15]
Gale, D.K.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. Adv. Funct.
Mater. 2009, 19, 926-933.
[16]
Jeffryes, C; Solanki, R.; Rangineni, Y.; Wang, W.; Chang, C.-H.;
Rorrer, G.L. Adv. Mater. 2008, 20, 2633-2637.
[17]
Wang, W.; Gutu, T.; Gale, D.K.; Jiao, J.; Rorrer, G.L.; Chang, C.-H. J.
Am. Chem. Soc. 2009, 131, 4178-4179.
[18]
Losic, D.; Mitchell, J.G.; Voelcker, N.H. Adv. Mater. 2009, 21, 29472958.
100
[19]
Yang, W.; Lopez, P.J.; Rosengarten, G. Analyst 2011, 136, 42-53.
[20]
Nassif, N.; Livage, J. Chem. Soc. Rev. 2011, 40, 849-859.
[21]
Jeffryes, C.; Gutu, T.; Jiao, J.; Rorrer, G.L. Mat. Sci. Eng. C 2008, 28,
107-118.
[22]
Qin, T.; Gutu,T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. ACS Nano 2008,
2, 1296-1304.
[23]
Harrison, P.J.; Waters, R.E.; Taylor, F.J.R. J. Phycol. 1980, 16, 25-35.
[24]
Hildebrand, M.; Frigeri, L.G.; Davis, A.K. J. Phycol. 2007, 43, 730740.
[25]
Brzezinski, M.A.; Conley, D.J. J. Phycol. 1994, 30, 45-55.
[26]
Kröger, N.; Poulsen, N. Annu. Rev. Genet. 2008, 42, 83-107.
[27]
Zurzola, C.; Bowler, C. Plant Physiol. 2001, 127, 1339-1345.
[28]
Jeffryes, C.; Gutu, T.; Jiao, J.; Rorrer, G.L. ACS Nano 2008, 2, 21032112.
[29]
Humbert, B. J. Non-Cryst. Solids 1995, 191, 29-37.
[30]
Riegel, B.; Hartmann, I.; Kiefer,W.; Groß, J.; Fricke, J. J. Non-Cryst.
Solids 1997, 211, 294-298.
[31]
Yuan, P.; He, H.P.; Wu,D.Q.; Wang, D.Q.; Chen, L.J. Spectrochim.
Acta. A 2004, 60, 2941-2945.
[32]
Blin, J.L.; Carteret, C. J. Phys. Chem. C. 2007, 111, 14380-14388.
[33]
Kammer, M.; Hedrich, R.; Ehrlich, H.; Popp, J.; Brunner, E.; Krafft, C.
Anal. Bioanal. Chem. 2010, 398, 509-517.
[34]
Durben, D.J.; Wolf, G.H. Phys. Rev. B. 1991, 43, 2355-2363.
101
[35]
Galeener, F.L.; Mikkelsen, J.C.; Geils, R.H.; Mosby, W.J. Appl. Phys.
Lett. 1978, 32, 34-36.
[36]
Micoulaut, M.; Cormier, L.; Henderson, G.S. J. Phys.-Condens. Mat.
2006, 18, R753-R784.
[37]
Céreyon, A.; Champagnon, B.; Martinez, V.; Maksimov, L.; Yanush,
O.; Bogdanov, V.N. Opt. Mater. 2006, 28, 1301-1304.
[38]
Townley, H.E.; Woon, K.L.; Payne, F.P.; White-Cooper, H.; Parker,
A.R. Nanotechnology 2007, 18, 295101.
[39]
Gale, D.K.; Jeffryes, C.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L.
J. Mater. Chem. 2011, doi: 10.1039/c1jm10861a.
[40]
Martin-Jézéquel, V.; Hildebrand, M.; Brzezinski, M.A. J. Phycol. 2000,
36, 821-840.
[41]
Brzezinski, M.A.; Olson, R.J.; Chisholm, S.W. Mar. Ecol. Prog. Ser.
1990, 67, 83-96.
[42]
Kaczmarska, I.; Beaton, M.; Benoit, A.C.; Medlin, L.K. J. Phycol.
2005, 42, 121-138.
[43]
Tesson, B.; Hildebrand, M. J. Struct. Biol. 2010, 169, 62-74.
[44]
Sumper, M.; Kröger, N. J. Mater. Chem. 2004, 14, 2059-2065.
[45]
Hildebrand, M.; York, E.; Kelz, J.I.; Davis, A.K.; Frigeri, L.G.;
Allison, D.P.; Doktycz, M.J. J. Mater. Res. 2006, 21, 2689-2698.
[46]
McCreery, R.L. Raman Spectroscopy for Chemical Analysis; WileyInterscience: New York, 2000.
[47]
Wopenka, B.; Pasteris, J.D. Anal. Chem. 1987, 59, 2165-2170.
[48]
White, S.N. Chem. Geol. 2009, 259, 240-252.
102
0.8
IIB
IIA
4.E+06
0.6
cell density
Si
3.E+06
(a)
0.2
1.E+06
0.E+00
0.0
0
Cell Number Density (# cells/mL)
0.4
2.E+06
Silicon Concentration (mmol Si/L)
Stage I
80
160 240 320
Cultivation Time (hr)
400
0.80
4.E+06
cell density
Si
0.60
3.E+06
(b)
2.E+06
0.40
0.20
1.E+06
0.E+00
Silica Concentration, CSi (mM)
Cell Number Density (# cells/mL)
5.E+06
0.00
0
80
160
240
320
400
Culture Time (hrs)
0.8
cell density
Si
4.E+06
0.6
(c)
3.E+06
0.4
2.E+06
0.2
1.E+06
0.E+00
Silicon Concentration (mmol Si/L)
Cell Number Density (# cells/mL)
5.E+06
0.0
0
80
160 240 320
Cultivation Time (hr)
400
Figure 4-1. (a) Cell number density and soluble Si concentration in the
culture medium of Stage II for a 48 hour addition of Si and Ge. (b)
Cell number density and soluble Si concentration in the culture medium
of Stage II for a 24 hour addition of Si and Ge. (c) Cell number density
and soluble Si concentration in the culture medium of Stage II for a 48
hour addition of Si.
103
24
Si
Si no uptake
Ge
Ge no uptake
1.2
20
1.0
(a)
0.8
16
12
0.6
8
0.4
4
0.2
0.0
Ge Concentration (μmol Ge/L)
Silicon Concentration (mmol Si/L)
1.4
0
-20
0
20
40
60
80 100 120
Stage II Cultivation Time (hr)
24
1.0
20
0.8
16
Si
0.6
Si no uptake
Ge
0.4
(b)
Ge no uptake
12
8
0.2
4
0.0
Germanium Concentration (µM)
Silica Concentration (mM)
1.2
0
-20 0
20 40 60 80 100 120 140 160 180
Culture Time (hrs)
Silicon Concentration (mmol Si/L)
1.4
1.2
1.0
Si
Si no uptake
0.8
0.6
(c)
0.4
0.2
0.0
-20
0
20
40
60
80 100 120
Stage II Cultivation Time (hr)
Figure 4-2. Si and Ge concentration profile in photobioreactor medium
during Stage II and concentration profile in the culture medium if there
was no uptake. (a) Profile for 48 hour Si and Ge addition in Stage II
which corresponds to Figure 1a. (b) Profile for 24 hour Si and Ge
addition in Stage II which corresponds to Figure 1b. (c) Profile for 48
hour Si addition in Stage II which corresponds to Figure 1c.
104
Hypovalve
formation
S
4.E+06
Cell
division
G2 + M
0.6
Ge enrichment in
the girdle band
0.5
G1
0.4
cell density
Ge
(a)
3.E+06
0.3
0.2
Ge enriched in girdle band
2.E+06
0.1
Stage IIA
Stage IIB
1.E+06
Ge Weight in the biosilca
Cell Number Density (# cells/mL)
5.E+06
0.0
0
20
40
60
80
Stage II Cultivation Time (hr)
100
120
Mass Ratio of Ge to Si in the Frustule
(g Ge / g Si)
0.50
0.40
(b)
0.30
0.20
Valve
0.10
Girdle Band
0.00
0
20
40
60
80
100
Stage II Cultivation time (hr)
120
Figure 4-3. (a) A schematic of the diatom cell cycle which correlates to the
cell density and Ge uptake as a function of cultivation time. (b)
Predicted Ge/Si mass ratio in the frustule valve and girdle band as a
function of cultivation time.
.
105
Figure 4-4. STEM-EDS spot scan analysis of an intact biosilica frustule
valve for the detection of Ge. (a) Spot denoted with a 1 indicates the
location of analysis. (b) EDS energy spectrum which confirmed the
metabolic insertion of Ge with the Kα energy peak at 9.86 keV.
106
(a)
1 µm
1 µm
(b)
200 nm
0.2 µm
Figure 4-5. TEM images of an (a) intact biosilica frustule and the (b)
submicron patterned pore structure of the frustule valve obtained at the
end of Stage I prior to Ge and Si addition in Stage II.
107
(b)
(a)
1 µm
250 nm
1 µm
0.5 µm
(c)
(d)
0.250
5 µm
nm
100 nm
100 nm
Figure 4-6. TEM images of an (a) intact biosilica frustule,(b),(c) submicron
patterned pore structure, and (d) porous nanostructure of the frustule
valve obtained at 120 hours into Stage II addition of Si and Ge.
108
(b)
(a)
hV = 1.05 µm
epivalve
dV = 9 µm
epitheca
epicingulum ( parent girdle bands)
hypocingulum (daughter girdle bands)
dGB = 9 µm
frustule
tf = 0.20 µm
hGB = 4.5 µm
hypotheca
hypovalve
1 µm
(c)
girdle bands
Si/Ge
valve
Stage IIA
Stage IIB
Figure 4-7. (a) SEM image of a Cyclotella frustule which shows the parent
valve (epivalve), parent girdle band (epicingulum), daughter valve
(hypovalve), and daughter girdle band (hypocingulum). (b) Schematic
of a diatom frustule which shows the dimensions used for model
development. (c) Schematic of the diatom frustule hypotheca at the
end of Stages IIA and IIB which shows Ge enrichment in the daughter
girdle bands.
Raman Signal Intensity (counts)
109
(a)
300
200
100
0
350
450
550
650
750
850
950 1050
Raman Shift (cm-1)
Normalized Raman Signal Intensity
1
(b)
0.8
0.6
0.4
0.2
0
360
400
440
480
520
560
Normalized Raman Signal Intensity
Raman Shift (cm-1)
1.0
(c)
0.8
0.6
0.4
0.2
0.0
360
400
440
480
Raman Shift (cm-1)
520
560
Figure 4-8. (a) Raman spectrum of frustule biosilica doped with Ge. (b)
Raman spectrum of Ge doped biosilica showing SiO2 peak (490 cm-1)
and GeO2 peak (420 cm-1). (c) Raman spectrum of biosilica without
Ge.
110
1.2
Raman Signal Intensity Ratio Ge:Si
Raman Signal Intensity Ratio Ge:Si
0.8
(a)
0.6
0.4
0.2
(b)
0.8
0.4
0.0
0.0
0
1
2
3
4
5
6
7
8
0
9
1
2
3
Distance (µm)
8
9
10
1.2
1.2
(c)
1
10
0
0.8
Raman Signal Intensity Ratio Ge:Si
Raman Signal Intensity Ratio Ge:Si
4 5 6 7
Distance (µm)
2 µm
0.6
0.4
0.2
0
0
1
2
3
4
5
6
Distance (um)
7
8
9
( d)
1
0.8
0.6
0.4
0.2
0
0
1
2
3
4
5
6
7
8
9
10 11
Distance (um)
Figure 4-9. Representative Raman line scans of normalized Raman intensity of
GeO2 (420 cm-1) to SiO2 (970 cm-1). (a), (b) Ge doped biosilica. (c)
Frustule valve and girdle band from Stage II fed Ge, which did not
have Ge. (d) Frustule from Stage II which was only fed Si. (c-inset)
Representative biosilica frustule for Raman spectroscopy analysis.
111
-16
(a)
Y (µm)
-14
-12
-10
-8
-16
0.5 µm
1 µm
5
(b)
10
X (µm)
15
Y (µm)
-14
-12
-10
-8
-16
0.5 µm
1 µm
5
(c)
10
X (µm)
15
Y (µm)
-14
-12
-10
0.5 µm
1 µm
-8
5
10
X (µm)
15
Figure 4-10. Raman map of intact Cyclotella sp. frustule. (a) Blue and (b) red
set to the silica raman bands of 970 cm-1 and 490 cm-1. (c) GeO2 Raman
band of 420 cm-1 set to green.
112
Table 1. Process parameters for two-stage photobioreactor
cultivation of Cyclotella sp.
Process parameter
Initial Si concentration (mM)
Final cell number density (cells/mL)
Si and Ge Stage II Delivery
Time of addition (hr)
Delivery feed rate (mL/hr)
Si molar flowrate (µmol Si/hr)
Ge molar flowrate (µmol Ge/hr)
Growth parameters
Cell yield (109 cells/mmol Si)
Specific growth rate, (h-1)
Stage
I
I
II
Control
0.69 ± 0.02
2.25·106 ± 5.68·104
4.87·106 ± 1.87·104
Ge
0.70 ± 0.03
1.79·106 ± 4.00·103
3.78·106 ± 5.60·104
II
II
II
II
48
3
94
0
48
3
94
1.4
I
II
I
II
2.45 ± 0.17
1.87 ± 0.04
0.013 ± 0.001
0.010 ± 0.001
3.80 ± 0.21
2.16 ± 0.13
0.013 ± 0.001
0.012 ± 0.003
113
Chapter 5
Immunocomplex Detection by Photoluminescent Germanium Centers
Doped in Diatom Biosilica
Debra K. Gale,*a Timothy Gutu,b Jun Jiao,b Chih-Hung Chang,a, Gregory L.
Rorrera
a
School of Chemical, Biological, and Environmental Engineering
Oregon State University, Corvallis, Oregon 97330 USA
b
Department of Physics
Portland State University, Portland, Oregon 97207 USA
114
ABSTRACT
Diatoms are single celled photosynthetic algae which fabricate silica
shells, called frustules, which possess a unique patterned pore array within a
micro-structure morphology. In this study, germanium (Ge) doped diatom
biosilica reports antibody and immunocomplex formation on the frustule
biosilica surface by an enhancement in the photoluminescence. Diatoms were
metabolically doped with Ge, up to 0.4 weight percent, with a two stage
photobioreactor cultivation strategy which targeted the Ge into the girdle band
of the Cyclotella sp. frustule. Upon thermal annealing in air at 400°C to create
photoluminescent germanium oxide centers, the baseline photoluminescence
signal was enhanced by up to 10 times. The Ge doped, annealed biosilica
reported antibody functionalization on the surface by up to a 50 times
enhancement in the PL, although it was unable to report immunocomplex
formation by a further enhancement in the PL. The surface antibody
concentration was reduced, which tuned the PL signal to report
immunocomplex formation on the diatom biosilica surface. The Ge doped
biosilica reported immunocomplex formation with a PL signal which was
almost 3 times higher than the PL signal of the native biosilica without Ge
upon immunocomplex formation. This is the first study to demonstrate that
the baseline diatom biosilica PL is a tunable parameter for PL based sensor
applications.
KEYWORDS
Diatom; Biosilica; Germanium; Immunocomplex; Photoluminescence
115
INTRODUCTION
There is significant interest in using silicon based nano- and microparticles, films and meso-structures as transducer platforms for biosensors. [14] Silica can be readily functionalized due to its surface chemistry, it is
biologically benign, non-toxic, water-stable, and can have a large surface area.
Sensors devices which employ a silica platform have used capacitance, [5, 6]
resistance, [7, 8] optical reflectivity, and photoluminescence (PL) [9-11] as the
detection transducer. PL based detection is a promising detection method
because it inherently exhibits a low signal background. [11] However, not all
morphologies of silica are photoluminescent. Silica nano- and micro-particles
must be doped with a luminophore upon synthesis, or porous silicon must be
ultrasonicated into smaller, typically non-uniform pieces to yield
photoluminescent particles. [2, 12-14] Films are made photoluminescent by
etching crystalline silicon wafers in hydrofluoric acid and depending on the
reaction conditions, the morphology and photoluminescence can vary greatly.
[15] Mesoporous silica is intrinsically photoluminescent, however the
synthesis conditions can significantly change the photoluminescence, [16] and
often it is doped with dyes, tin, and other luminescent materials to increase the
PL intensity. [17, 18]
Photoluminescent silica can also be fabricated biologically by diatoms.
[19-22]
Diatoms are single celled photosynthetic algae which fabricate silica
shells, called frustules, which possess a unique patterned pore array within a
microstructure morphology. Diatom PL, like that of nanostructured silica,
derives from surface defects such as silanol groups. [19, 22] Biological
fabrication of silica by diatoms has promising potential as a miniaturized PL
based transducer of biomolecule and gas recognition events. [23] Recently, we
showed that diatom biosilica was able to report immunocomplex formation on
the biosilica surface by a 3 times enhancement in the PL intensity of
116
immunocomplex functionalized biosilica above that of antibody functionalized
biosilica. [21] It has also been shown that diatom biosilica PL changes upon
covalent functionalization with a peptide [24] or interaction with NO2 gas. [25,
26] In these studies the change in the photoluminescence was induced by
interaction of ligands with the photoluminescent surface silanol groups on the
diatom biosilica.
In this study, we used photoluminescent germanium (Ge) centers
metabolically doped into the diatom frustule girdle band to report antibody and
immunocomplex functionalization on the biosilica surface. Ge centers
biologically inserted into the frustule biosilica enhanced the intrinsic
photoluminescence intensity of the biosilica, which increased the detection
signal of an antibody and immunocomplex. Diatom frustules with luminescent
Ge centers for PL based detection of immunocomplex functionalization can be
fabricated on a massively parallel and reproducible scale, devoid of challenges
associated with conventional fabrication of photoluminescent silica.
EXPERIMENTAL
Two Stage Diatom Cell Cultivation
Axenic cultures of the photosynthetic marine diatom Cyclotella sp. were
obtained from UTEX The Culture Collection of Algae (#1269). Maintenance
culture of this organism in Harrison’s Artificial Seawater Medium [27] was
previously described. [21] A two-stage photobioreactor cultivation strategy
was developed to metabolically dope Ge into the girdle band of the Cyclotella
sp. diatom frustule. [28] Cultivations were carried out at 150 µE m-1s-1
incident light intensity, 14 hour light/10 hour dark photoperiod, 0.10 air min-1
aeration rate (~350 ppm CO2), and a constant temperature of 22ºC. In stage I
of the cultivation diatom cells were inoculated into 4.5 L of artificial seawater
medium to a cell density of 2.0x105 cell/mL with 0.65 mM soluble silicon
117
(Na2SiO3) and grown to silicon (Si) starvation and stationary phase. In stage II
of the cultivation, a soluble mixture of 31.25 mM Si and 0.47 mM Ge in
destilled/deionized water, from Na2SiO3 and GeO2, was delivered into the
photobioreactor medium, controlled by a syringe pump, at a volumetric flow
rate of 3 mL hr-1 for 2 photoperiods (48 hours) which began at the start of the
light phase, to target one cell division. The control cultivation was carried out
using the same parameter previously described, except only Si was fed to the
cultivation in stage II. The photobioreactor cultivation was monitored by
measurements of the cell number density in triplicate with a Beckman Z2
Coulter Counter and spectrophotometric measurement of Si and Ge in the
culture medium as described previously. [22] Biosilica frustules extracted
from diatom cells at 0, 24, 38, 48, 72, and 120 hours into stage II contained
0.0132 ± 0.0058, 0.1681 ± 0.0058, 0.0092 ± 0.0058, 0.2443 ± 0.0049, 0.2801 ±
0.0310, 0.3245 ± 0.0046, and 0.4075 ± 0.0334 weight percent Ge in silica as
measured by inductively coupled plasma (ICP) described previously. [29, 30]
Frustule Isolation and Electron Microscopy
Biosilica frustule isolation from the Cyclotella sp. cells and electron
microscopy imaging conditions were previously described. [21] Frustules
were isolated from the diatom cells by treatment with 30 wt. % aqueous
hydrogen peroxide at ph 2.5, which oxidized organic cell material. Frustules
isolated from diatom cells were suspended in methanol, then deposited onto a
holey carbon coated copper TEM grid and allowed to air dry. Frustules were
then imaged by transmission electron microscopy (TEM) or scanning electron
microscopy (SEM).
Thermal Annealing of Diatom Biosilica
118
Biosilica frustules which were isolated from the diatom cells by hydrogen
peroxide treatment were thermally annealed in air in a pre-heated furnace at
400ºC for 2 hours as previously described. [22] The PL was measured before
and after annealing of the same material which had been allowed to cool for at
least one hour to reach room temperature.
Diatom Frustule Functionalization
Functionalization of the native diatom frustule biosilica (without Ge) with
Rabbit IgG antibody and anti-Rabbit IgG antigen was previously described in
detail. [21] The surface of the diatom biosilica (10 mg of frustule mass) was
functionalized with amine groups by reaction of 14 mM 3aminopropyltrimethoxysilane (APS) or 2.8 µM APS in 2 mL of ethanol.
Aminated biosilica (40 µg) was deposited on a circular glass coverslip before
antibody functionalization in individual wells of a 6-well polystyrene plate to
facilitate photoluminescence spectroscopy measurements. Aminated biosilica
was covalently attached to Rabbit IgG antibody (Pierce Biotechnology #31235,
0.0234 mg mL-1) in phosphate buffered saline, pH 7.2 with BS3 (bissulfosuccinimidyl suberate, Pierce Biotechnology # 21580, 0.574 µmol g-1
biosilica-APS) an amine reactive homobifunctional cross linker. The antibody
functionalized biosilica was then functionalized with the complimentary
antigen, goat anti-Rabbit IgG (Pierce Biotechnology # 31210, 0.0797 mg L-1).
After functionalization, the films were dip rinsed in PBS buffer to remove
excess antigen, then allowed to air dry for photoluminescence measurements.
Samples prepared for epifluorescence microscopy were functionalized with
fluorescein labeled goat anti-Rabbit IgG (Pierce Biotechnology #31635) and
imaged under a 470 nm excitation source and GFP filter. All samples were
prepared in triplicate.
119
Photoluminescence Microscopy
The biosilica films described above were excited with a 337 nm N2 gas laser
source (Spectra Physics VSL, 30 kW peak power, 2.4 mW average power, 20
Hz, 4 ns pulse, 120 µJ) as previously described. [21, 22] The emission
spectrum was collected with an Acton Inspectrum 300 spectrometer equipped
with a CCD detector (0.20 mm slit width, 300 gratings per millimeter, 2
second integration time). The PL of the same material was measured after
selected steps in the functionalization process to facilitate PL intensity
normalization to the bare biosilica.
RESULTS
Ge enrichment in the girdle band of the frustule
Ge was metabolically doped into the girdle band of the Cyclotella frustule
using a two stage photobioreactor cultivation strategy we developed
previously. [28] From the valve face, the diatom frustule appeared as a silica
disk encircled with an enriched Ge ring. In stage I of the cultivation strategy
cells of the diatom Cyclotella sp. were grown up to silicon (Si) starvation and
stationary phase. TEM and SEM images of frustules which were isolated from
diatom cells at the end of stage I are presented in Figure 2a and 2b. Cyclotella
is a centric diatom which is comprised of two disk shaped valves,
approximately 10 µm in diameter, that are connected with a ring shaped girdle
band, approximately 6 µm in thickness. The valve is made up of radially
symmetrical patterned pores 100 nm in diameter, Figure 2a, which are filled
with 4-5 pores approximately 20 nm in diameter, Figure 2b.
In stage II of the cultivation, the cells were fed a solution of Si and
germanium (Ge) for two photoperiods, indicated by the dotted lines in Figures
1a and 1b, to allow for one cell division. The concentration profile of Si and
Ge in the photobioreactor medium and the concentration if there was no uptake
120
by the cells is presented in Figure 1b. TEM images of frustules extracted from
cells 120 hours into stage II of the cultivation which was fed Ge, are presented
in Figure 2c and 2d. Ge insertion into the frustule biosilica appeared to alter
the pore structure. The morphology of the pores which extended from the
valve center appeared to decrease in diameter. Furthermore, the four to five 20
nm in diameter pores which were set within the 100 nm pores were not present
in the frustule doped with Ge, as shown in Figures 2b and 2d. Attempts to
image the girdle band structure with TEM, were unsuccessful due to the
thickness of the material and lack of electron transparency. Biosilica frustules
extracted from diatom cells at 24, 38, 48, 72, and 120 hours into stage II
contained 0.1681 ± 0.0092, 0.2443 ± 0.0049, 0.2801 ± 0.031, 0.3245 ± 0.0046,
and 0.4075 ± 0.0334 weight percent Ge in silica.
Enhanced immunocomplex detection by Photoluminescence of Ge doped
frustule biosilica
Diatom biosilica metabolically doped with Ge in the girdle band of the
frustule in stage II was annealed at 400°C in air for two hours to enhance the
intrinsic photoluminescence (PL) of the biosilica. [22] Previously we showed
that Ge metabolically doped in Pinnularia sp. diatom biosilica was primarily in
the form of GeO2 and could be thermally converted to intensely
photoluminescent GeO. [22] The annealed biosilica was then tested for its
ability to report antibody and immunocomplex functionalization on the surface
of the biosilica by a further enhancement in the PL.
The PL of the biosilica was enhanced by at least 10 times after thermal
annealing in air for 2 hours, for biosilica sampled between 24 and 72 hrs in
stage II, and slightly less than that for biosilica extracted at 120 hours, as
shown in Figure 3. The biosilica extracted at 0 hours into stage II, which did
not contain Ge only increased by approximately a factor of two after annealing
121
in air. The PL signal in Figure 3 is normalized to the PL signal of the biosilica
prior to annealing in air. Thermal annealing of the Ge doped biosilica in air at
400°C for two hours resulted in an intense enhancement of the baseline PL
signal intensity.
The annealed Ge doped biosilica was tested for the ability to report
antibody functionalization on the surface by an increase in the PL signal
intensity. The biosilica was functionalized with Rabbit IgG antibody, which
was reported by a secondary intense enhancement in the PL signal intensity as
presented in Figure 3. The PL signal intensity of the annealed-antibody
functionalized biosilica was normalized to the signal intensity of the biosilica
prior to annealing, the same normalization as the annealed biosilica. Only the
biosilica doped with Ge reported antibody functionalization by an
enhancement of over 30 times the signal intensity compared to the biosilica
before annealing. The biosilica without Ge also reported antibody
functionalization by an increase of the PL signal by approximately a factor of
3, however this signal intensity enhancement was 10 times lower than that of
the Ge doped biosilica.
Epifluorescence microscopy was used to confirm the presence of
immunocomplex formation on the annealed biosilica surface. Biosilica
frustules acquired at 120 hours into stage II of the cultivation were annealed at
400°C for two hours then functionalized with an antibody. The annealedantibody functionalized biosilica was then challenged with goat anti-Rabbit
IgG which was labeled with fluorescein. An epifluorescence image of a
fluorescein labeled-immunocomplex functionalized frustule is presented in
Figure 4. Previously, we showed that the antibody and antigen were not
simply adsorbed to the surface of the biosilica but covalently functionalized to
the biosilica. [21] Epifluorescence microscopy confirmed that the annealed
biosilica was readily functionalized. The contoured surface of Cyclotella is
122
evident from the scalloped perimeter of the frustule and although the patterned
porous surface is beyond the visible diffraction limit, the fultoportulae, which
is the large pore located in the center of the valve in Figure 2c, is visible as an
intense green spot off center from the middle of the valve.
The Ge doped, annealed-antibody functionalized biosilica was then
tested for the ability to detect immunocomplex formation by a PL
enhancement. The antibody-functionalized, Ge doped, annealed biosilica did
not report immunocomplex functionalization on the surface by enhancement in
the PL. Presented in Figure 5, is the normalized PL signal intensity of Ge
doped biosilica extracted from 120 hours into the stage II cultivation which
was annealed and functionalized with an antibody and an antigen. The
normalized PL intensity of biosilica with 0.4 wt.% Ge which was
functionalized with an immunocomplex under 14 mM APS reaction conditions
is a representative result for all of the biosilica extracted in stage II which was
doped with Ge. Previously, we showed that native biosilica, without Ge,
functionalized with these APS reaction conditions, reported immunocomplex
formation by a 2 times enhancement in the PL signal of the antibody
functionalized biosilica. [21] There was not a PL signal intensity enhancement
of the Ge doped, annealed-antibody functionalized biosilica after
immunocomplex formation. Furthermore, the PL intensity was slightly
quenched. The reaction conditions of the amination step were diluted by a
factor of 5000 to reduce the number of surface amination sites, which likely
reduced the surface antibody concentration. Reduction of the surface amine
concentration on the biosilica surface likely reduced the biosilica surface
antibody concentration such that discrimination in the PL signal was observed
between antibody and immunocomplex formation, as shown in Figure 5. The
biosilica doped with 0.4 wt. % Ge reported a more intense PL signal upon
antibody and immunocomplex formation than the biosilica without Ge for the
123
diluted 28.6 µM APS reaction conditions, as shown in Figure 5. The PL
spectra of the native biosilica without Ge and the Ge doped biosilica
functionalized with an antibody and immunocomplex with diluted APS
concentrations is presented in Figures 6a and 6b. The PL signals are
normalized to the PL of the biosilica prior to annealing treatment. The spectra
show that the peak height intensity wavelength for the annealed biosilica and
functionalized biosilica remained around 450 nm. The Ge doped biosilica
reported a higher PL signal upon immunocomplex functionalization that the
native biosilica without Ge.
DISCUSSION
This study has shown that photoluminescent GeO centers metabolically
doped in diatom biosilica can be used as a PL based transducer to report
antibody and immunocomplex formation. Previous work in our laboratory
showed that selective and quantitative immunocomplex formation on an
antibody-functionalized native biosilica (no Ge) surface could be reported by
an enhanced PL signal. [21] This study is unique because we showed that that
GeO centers doped in Cylclotella diatom biosilica are up to 15 times more
photoluminescent than surface silanols (Figure 3) of native biosilica, and it
was hypothesized that if the baseline PL signal of the biosilica was increased,
then the PL signal which reported antibody and immunocomplex
functionalization would proportionately be increased.
Previously, we showed that diatom biosilica emits blue PL upon
excitation with 337 nm light. [19, 21] The blue PL of diatom biosilica
originates from surface defects as silanol (SiOH) groups, [19, 21, 31] like that
of nanostructured silica. [32-36] Work also done in our laboratory showed
that, like Ge doped silica made by conventional fabrication methods, [37-40]
diatom biosilica metabolically doped with Ge exhibited an amplified PL signal
124
attributed to germanium oxide (GeO) defects. [22] By annealing biosilica at
400°C, the metabolically doped germanium dioxide (GeO2) was thermally
converted to intensely photoluminescent GeO centers, and the intrinsic
luminescent SiOH groups of the native biosilica were quenched, which
resulted in a diatom frustule structure which emitted PL from GeO luminescent
centers. [22]
This is the first study to biologically amplify the intrinsic
photoluminescence properties of biosilica for enhanced antibody and
immunocomplex detection by PL. Recently, there has been significant interest
in attaching various biomolecules such as antibodies and proteins to the surface
of diatom frustule biosilica. [21, 24, 41-43] De Stefano et al., [24] like our
previous study [21] observed a change in the PL upon biomolecule
functionalization. However, to date there have been no reports on modification
of the intrinsic baseline PL of the diatom biosilica to tune the
photoluminescence detection signal to report functionalization. The Ge doped
biosilica reported antibody functionalization by a PL enhancement by a factor
of up to 55 compared to the biosilica without Ge which only reported antibody
functionalization by a 3 times enhancement in the PL, as presented in Figure
3. The PL signal of the Ge doped biosilica responded sensitively to antibody
functionalization, that upon immunocomplex formation the PL signal did not
respond, and in some cases was even slightly quenched as shown in Figure 5.
This observation confirms that the baseline PL of the biosilica is an important
parameter in reporting biomolecule detection but also holds potential as being
a tunable parameter for PL based sensing.
The APS reaction conditions were decreased to gain sensitivity in PL
detection between the antibody and immunocomplex. There were two reasons
to decrease the APS reaction conditions. First, by reduction of the APS
concentration, it was likely that the surface amine concentration was reduced
125
which also reduced the surface antibody concentration. Second, amine groups
are known to emit blue photoluminescence. [44, 45] Modification of the
surface APS concentration reduced the PL signal contribution from the amine
groups to the final PL signal of the antibody and immunocomplex
functionalized biosilica. The PL spectrum of native biosilica, 0 wt. % Ge and
biosilica doped with 0.4 wt. % Ge at the diluted APS reaction conditions upon
antibody and immunocomplex functionalization are presented in Figure 6a
and Figure 6b. At the diluted APS conditions, the Ge doped biosilica reported
antibody and immunocomplex formation with a signal 3 times higher than the
biosilica without Ge. However, the selectivity, which is defined as the ratio of
the immunocomplex PL signal intensity to the antibody PL signal intensity, of
the biosilica without Ge is greater than the Ge doped biosilica. The Ge doped
biosilica reports a more intense PL signal intensity upon immunocomplex
formation compared to the native biosilica, even though the native biosilica
exhibits higher immunocomplex selectivity.
Previously we showed that annealing the biosilica at 400ºC in air for 2
hours quenched the intrinsic PL centers of the native biosilica. [22] The PL
signal in Figures 5 and 6 of the Ge doped biosilica is made up of a weak PL
signal from quenched surface silanols and an intense PL signal from GeO
photoluminescent centers. Since the diatoms were cultivated in such a way
that the Ge was spatially enriched in the girdle band of the frustule, the PL is
likely emitted primarily from the girdle band. [28] From the valve face view,
the functionalized frustule likely appeared as a blue photoluminescent ring
upon excitation, even though the entire frustule was functionalized. In future
work, we will characterize spatial location of PL emission from the
functionalized biosilica.
GeO doped into semiconductor films can act as a sensitizer to increase
the PL emission. [46-49] In this study we showed that GeO photoluminescent
126
centers report a more intense PL signal upon surface functionalization than the
instrinsic surface silanols. Furthermore, nucleophilic biomolecules, such as an
antibody and immunocomplex, [50] attached to photoluminescent surfaces can
increase the PL emission. [44, 51-60] In our previous work [21] we suggested
that the nucleophilic immunocomplex functionalized on the native biosilica
(without Ge) increased the PL intensity by donating electrons to non-radiative
PL defects sites, which decreased non-radiative electron decay and increased
radiative electron decay. We propose that the luminescent GeO centers are
more sensitive to radiative recombination than luminescent silanols, which
would result in a higher PL signal to report immunocomplex formation.
CONCLUSIONS
In this study, we showed that photoluminescent GeO centers
metabolically doped in diatom biosilica can be used to report antibody
functionalization and immunocomplex formation by an intense enhancement in
the PL signal. Ge, up to 0.4 weight percent was metabolically doped into the
diatom biosilica by a two stage photobioreactor cultivation strategy which
enriched Ge in the girdle band of the frustule. GeO photoluminescent centers
metabolically doped in the biosilica were used to report antibody and
immunocomplex formation on the diatom surface. Biosilica doped with Ge
emitted a more intense PL signal to report immunocomplex formation than the
native biosilica without Ge. This is the first study to demonstrate that the
baseline diatom biosilica PL is a tunable parameter in PL based detection.
Furthermore, PL based detection is label free and contained within the
microstructured form factor of the diatom biosilica which is fabricated under
environmentally benign conditions.
ACKNOWLEDGEMENTS
127
This research was supported by the National Science Foundation (NSF)
award number BES-0400468.
REFERENCES
[1]
Wang, L.; Zhao, W.; Tan, W. Nano Res.2008, 1, 99-115.
[2]
Knopp, D.; Tang, D.; Niessner, R. Anal. Chim. Acta 2009, 647, 14-30.
[3]
Gu, Z.; Chen, X.-Y.; Shen, Q.-D.; Ge, H.-X.; Xu, H.-H. Polymer 2010,
51, 902-907.
[4]
Jane, A.; Dronov, R.; Hodges, A.; Voelcker, N.H. Trends Biotechnol.
2009, 27, 230-239.
[5]
Schechter, I.; Ben-Chorin, M.; Kux, A. Anal. Chem. 1995, 67, 37273732.
[6]
Zhang, Q.; Shin, Y.J.; Hua, F.; Saraf, L.V.; Matson, D.W. J. Nanosci.
Nantechnol. 2008, 8, 3008-3012.
[7]
Lewis, S.E.; DeBoer, J.R.; Gole, J.L.; Hesketh, P.J. Sens. Actuators, B
2005, 110, 54-56.
[8]
Yao, Z; Yang, M. Sens. Actuators, B 2006, 117, 93-98.
[9]
Ko, M.C.; Meyer, G.J. Chem. Mater. 1996, 8, 2686-2692.
[10]
Gulino, A.; Giuffrida, S.; Mineo, P.; Purrazzo, M.; Scamporrino, E.;
Ventimiglia, G.; van der Boom, M.E.; Fragalá, I. J. Phys. Chem. B
2006, 110, 16781-16786.
[11]
Sailor, M.J.; Wu, E.C. Adv. Funct. Mater. 2009, 19, 3195-3208.
[12]
Heinrich, J.L.; Curtis, C.L.; Credo, G.M.; Sailor, M.J.; Kavanagh, K.L.
Science 1992, 255, 66-68.
[13]
Bley, R.A.; Kauzlarich, S.M.; Davis, J.E.; Lee, H.W.H. Chem. Mater.
1996, 8, 1881-1888.
[14]
Santra, S.; Zhang, P.; Wang, K.; Tapec, R.; Tan, W. Anal. Chem. 2001,
73, 4988-4993.
128
[15]
Zhang, X.G. J. Electrochem. Soc. 2004, 151, C69-C80.
[16]
Anedda, A.; Carbonaro, C.M.; Clemente, F.; Corpino, R.; Ricci, P.C.
Mater. Sci. Eng.,C 2003, 23, 1073-1076.
[17]
Liu, Z.C.; Chen, H.R.; Huang, W.M.; Gu, J.L.; Bu, W.B.; Hua, Z.L.;
Shi, J.L. Microporous Mesoporous Mater. 2006, 89, 270-275.
[18]
Sokolov, I.; Volkov, D.O. J. Mater. Chem. 2010, 20, 4247-4250.
[19]
Qin, T.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. J. Nanosci.
Nanotechno. 2008, 8, 1-7.
[20]
Jeffryes, C; Solanki, R.; Rangineni, Y.; Wang, W.; Chang, C.-H.;
Rorrer, G.L. Adv. Mater. 2008, 20, 2633-2637.
[21]
Gale, D.K.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. Adv. Funct.
Mater. 2009, 19, 926-933.
[22]
Gale, D.K.; Jeffryes, C.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L.
J. Mater. Chem. 2011, 21, 10658-10665.
[23]
Yang, W.; Lopez, P.J.; Rosengarten, G. Analyst 2011, 136, 42.
[24]
De Stefano, L.; Rotiroti, L.; De Stefano, M.; Lamberti, A.; Lettieri, S.;
Setaro, A.; Maddalena, P. Biosens. Bioelectron., 2009, 24, 1580-1584.
[25]
Bismuto, A.; Setaro, A.; Maddalena, P.; De Stefano, L.; De Stefano, M.
Sens. Actuators, B 2008, 130, 396-399.
[26]
Lettieri, S.; Setaro, A.; De Stefano, L.; De Stefano, M.; Maddalena, P.
Adv. Funct. Mater. 2008, 18, 1257-1264.
[27]
Harrison, P.J.; Waters, R.E.; Taylor, F.J.R. J. Phycol. 1980, 16, 25-28.
[28]
Gale, D.K.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. Germanium
Enrichment in the Girdle Band of the Centric Diatom Cyclotella sp.,
2011, manuscript in preparation.
[29]
Jeffryes, C.; Gutu, T.; Jiao, J.; Rorrer, G.L. Mat. Sci. Eng. C 2008, 28,
107-118.
129
[30]
Jeffryes, C.; Gutu, T.; Jiao, J.; Rorrer, G.L. ACS Nano 2008, 2, 21032112.
[31]
Qin, T.; Gutu,T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. ACS Nano 2008,
2, 1296-1304.
[32]
Rückschloss, M.; Wirschem, T.; Tamura, H.; Ruhl, G.; Oswald, J.;
Vep_rek, S. J. Lumin., 1995, 63, 279–287.
[33]
Garcia, M.A.; Paje, S.E.; Villagas, M.A.; Llopis, J. Mater. Lett. 2000,
43, 23-26.
[34]
He, N.; Ge, S.; Yang, C.; Hu, C., Gu, M. Mater. Res. Bull.2004, 39,
1931-1937.
[35]
He, N.Y.; Ge, S.X.; Yang, C.; Cao, J.M.; Gu, M. Mater. Lett., 2004, 58,
3304–3307.
[36]
Carbonaro, C.M.; Clemente, F.; Corpino, R.; Ricci, P.C.; Anedda, A. J.
Phys. Chem. B, 2005, 109, 14441–14444.
[37]
Gao, T.; Bao, X.M.; Yan, F.; Tong, S. Phys. Lett. A, 1997, 232, 321–
325.
[38]
Rebohle, L.; von Borany, J.; Gr€otzschel, R.; Markwitz, A.; Schmidt,
B.; Tyschenko, I.E.; Skorupa, W.; Fr€ob, H.; Leo, K. Phys. Status
Solidi A, 1998, 165, 31–35.
[39]
Li, J.; Wu, X.L.; Yang, Y.M.; Yang, X.; Bao, X.M. Phys. Lett. A, 2003,
314, 299–303.
[40]
Xu, Z.W.; Ngan, A.H.W.; Hua, W.Y.; Meng, X.K. Appl. Phys. A:
Mater. Sci. Process., 2005, 81, 459–463.
[41]
Townley, H.E.; Woon, K.L.; Payne, F.P.; White-Cooper, H.; Parker,
A.R. Nanotechnology 2007, 18, 295101.
[42]
De Stefano, L.; Lamberti, A.; Rotiroti, L.; De Stefano, M. Acta
Biomater. 2008, 4, 126-130.
130
[43]
Lin, K.-C.; Kunduru, V.; Bothara, M.; Rege, K.; Prasad, S.;
Ramakrishna, B.L. 2010, 25, 2336-2342.
[44]
Carlos, L. D.; Sa´ Ferreira, R. A.; Pereira, R. N.; Assunca˜o, M.; de Zea
Bermudez, V. J. Phys. Chem. B 2004, 108, 14924.
[45]
Bekiari, V.; Lianos, P. J. Nanosci. Nanotechnol. 2006, 6, 372-376.
[46]
Wu, J.; Punchaipetch, P.; Wallace, R.M.; Coffer, J.L. Adv. Mater. 2004,
16, 1444-1448.
[47]
Wu, J.; Coffer, J.L. Chem. Mater. 2007, 19, 6266-6276.
[48]
Wu, J.; Coffer, J.L.; Wang, Y.; Schulze, R. J. Phys. Chem. C 2009,
113, 12-16.
[49]
Kanjilal, A; Rebohle, L.; Voelskow, M.; Skorupa, W.; Helm, M. Appl.
Phys. Lett. 2009, 94, 051903.
[50]
Paul, S.; Nishiyama, Y.; Planque, S.; Karle, S.; Taguchi, H.; Hanson,
C.; Weksler, M. E. Springer Semin. Immun. 2005, 26, 485.
[51]
Dannhauser, T.; O’Neil, M.; Johansson, K.; Whitten, D.; McLendon,
G. J. Phys.Chem. 1986, 90, 6074.
[52]
Stewart, M.P; BuriakJ.M. Adv. Mater. 2000, 12, 859.
[53]
Jia, R. P.; Shen, Y.; Luo, H. Q.; Chen, X. G.; Hu, Z. D.; Xue, D. S.
Thin Solid Films 2005, 471, 264.
[54]
Liu, T.-Y.; Liao, H.-C.; Lin, C.-C.; Hu, S.-H.; Chen, S.-Y. Langmuir
2006, 22, 5804.
[55]
Wang, Q.; Kuo, Y.; Wang, Y.; Shin, G.; Ruengruglikit, C.; Huang, Q.
J. Phys.Chem. B. 2006, 110, 16860.
[56]
Dorfman, A.; Kumar, N.; Hahm, J. Adv. Mater. 2006, 18, 2685.
[57]
Dorfman, A.; Kumar, N.; Hahm, J., Langmuir 2006, 22, 4890.
[58]
Sarangi, S.N.; Goswami, K.; Sahu, S.N. Biosens. Bioelectron. 2007, 22,
3086.
131
[59]
Shang, Q.; Wang, H.; Yu, H.; Shan, G.; Yan, R. Colloid Surf. A 2007,
294, 86.
[60]
Wang, J.-H.; Wang, H.-Q.; Zhang, H.-L.; Li, X.-Q.; Hua, X.-F.; Huang,
Z.-L.; Zhao, Y.-D. Colloid Surf. A 2007, 305, 48.
[61]
Dybiec, M.; Chornokur, G.; Ostapenko, S.; Wolcott, A.; Zhang, J. Z.;
Zajac, A.; Phelan, C.; Sellers, T.; Gerion, D. Appl. Phys. Lett. 2007, 90,
263112.
132
Stage 2
Cell Number Density (# cells/mL)
0.8
cell density
Si
3.E+06
(a)
0.6
2.E+06
0.4
1.E+06
0.2
0.E+00
0.0
0
80
160
240
Silicon Concentration (mmol Si/L)
Stage I
4.E+06
320
25
1.4
Ge
Ge no uptake
Si
Si no uptake
20
1.2
1.0
(b)
15
0.8
0.6
10
0.4
5
0.2
0
Silicon Concentration (mmol Si/L)
Germanium Concentration (μmol Ge/L)
Cultivation Time (hr)
0.0
-20
0
20
40
60
80 100 120 140
Stage II Cultivation Time (hr)
Figure 5-1. (a) Cell number density and soluble Si concentration in the culture
medium of Stages I and II for a 48 hour addition of Si and Ge starting
at 131 hours indicated by the dotted line. (b) Si and Ge concentration
profile in the culture medium during Stage II and Si and Ge
concentration profile in the culture medium if there was no uptake.
133
(a)
1
1 µm
1 µm
(c)
1
1 µm
1 µm
(b)
1
200 nm
0.2 µm
(d)
1
200 nm
0.5 µm
Figure 5-2. TEM images of the valve facing view of an intact diatom frustule
and the submicron patterned pore detail on the frustule valve. Frustules
were extracted from Cyclotella diatom cells (a), (b) at the end of Stage
I prior to Si and Ge addition and (b), (c) 120 hrs into Stage II.
60
0.6
50
0.5
40
0.4
30
Ge biosilica 400°C + antibody
Ge biosilica 400°C
20
0.3
0.2
Ge wt. %
10
0.1
0
0.0
0
20
40
60
80
100
Stage II Cultivation Time (hr)
Ge wt. % in the Biosilica
Normalized Photoluminescence
Intensity
134
120
Figure 5-3. Normalized photoluminescence intensity of biosilica from Stage
II of the cultivation which was fed Ge and the Ge weight percent in the
biosilica. Biosilica was annealed in air at 400°C for 2 hours, then
functionalized with an antibody. The photoluminescence intensity is
normalized to the photoluminescence intensity of each sample prior to
annealing and functionalization treatment.
135
5 µm
Figure 5-4. Epifluorescence image of a diatom frustule acquired 120 hours
into Stage II of Si and Ge addition which was annealed at 400°C in air
for 2 hours then functionalized with a Rabbit IgG antibody and a
fluorescein labeled goat anti-Rabbit IgG antigen.
60
50
Biosilica 400°C
Biosilica 400°C + Antibody
Biosilica 400°C + Antibody + Antigen
Sensitivity Factor
2.8
2.4
2.0
40
1.6
30
1.2
20
0.8
10
0
0.4
Ge
00wt
wt %
% Ge
Ge 0.4
0.4wt
wt %
% Ge
Ge 0.4
0.4 wt
wt %
% Ge
APS
28.6
(0.36µM
molAPS
APS 28.6
(0.36µM
molAPS
APS 143
(1.8mM
mol APS
mol SiO2-1)
mol SiO2-1)
Immunocomplex Selectivity Factor
Normalized Photoluminescence Intensity
136
0.0
mol SiO2-1)
APS
Frustule APS:Biosilica Antibody:Biosilica Antigen:Biosilica
Concentration Mass (mg)
(mmol/g)
(µmol/g)
(µmol/g)
28.6 µM
143 mM
10
10
28.6
0.0057
0.52
0.52
17.7
17.7
Figure 5-5. Normalized photoluminescence intensity of biosilica with 0 and
0.4 wt. % Ge which was functionalized with an antibody and
immunocomplex with 28.6 µM (5000x dilution) or 143 mM (1x
dilution) APS concentration. Immunocomplex selectivity factor is
defined as the ratio of the immunocomplex PL signal intensity to the
antibody PL signal intensity.
137
Normalized Photoluminescence
Intensity
16
0.4 wt % Ge
14
biosilica 400°C +
antibody + antigen
12
10
biosilica 400°C
+ antibody
8
6
(a)
4
biosilica 400°C
2
0
350 400 450 500 550 600 650 700 750
Wavelength (nm)
Normalized Photoluminescence
Intensity
16
14
0 wt % Ge
(b)
12
10
8
6
4
2
biosilica 400°C +
antibody + antigen
biosilica 400°C +
antibody
biosilica 400°C
0
350 400 450 500 550 600 650 700 750
Wavelength (nm)
Figure 5-6. Normalized photoluminescence intensity of annealed biosilica
with (a) 0.4 and (b) 0 wt. % Ge which was functionalized with an
immunocomplex using a 28.6 µM APS concentration.
138
Chapter 6: Conclusion
Chapter Summary
In Chapter 2, diatom biosilica was functionalized with an antibody and
tested for immunocomplex detection by enhanced PL. Functionalization of the
diatom biosilica with the antibody amplified the intrinsic blue PL of the diatom
biosilica. Furthermore, the antibody-antigen immunocomplex increased the
peak PL intensity by at least a factor of two, whereas the non-complimentary
antigen did not. The PL emission was correlated to the antigen concentration
where immunocomplex binding followed a Langmuir isotherm with a binding
constant of 2.8 ± 0.7 x 10-7 M. We proposed that the nucleophilic
immunocomplex increased the PL intensity by donating electrons to nonradiative defect sites on the 138hotoluminescent diatom biosilica, thereby
decreasing non-radiative electron decay and increasing the radiative emission.
In Chapter 3, the PL of diatom biosilica metabolically doped with
germanium (Ge) was thermally activated to create a highly photoluminescent
nano- and micro-structured material. Controlled levels of Ge were doped into
the diatom biosilica frustule by a two stage photobioreactor cultivation
strategy. Germanium oxide (GeO) photoluminescent defects were enhanced
by thermally converting the germanium dioxide (GeO2) deposited into the
frustule by the diatom cell into highly photoluminescent GeO. Metabolic
doping of diatom biosilica with photoluminescent Ge defects amplified the
intrinsic baseline PL.
In chapter 4, a two stage photobioreactor cultivation strategy was
developed to dope Ge into the girdle band of the centric diatom cyclotella. In
stage I of the photobioreactor cultivation strategy, cells were grown to silicon
139
starvation and stationary phase. In stage II of the cultivation strategy, the cells
were co-fed a mixture of silicon and germanium for a period of 48 hours,
which was 2 photoperiods. The two stage photobioreactor cultivation strategy
was developed to take advantage of the cell cycle phases to direct germanium
deposition into the girdle band of the diatom frustule.
Micro Raman
spectroscopy mapping confirmed that germanium was enriched in the girdle
band by a factor of 3 compared to the valve. This experimental result agreed
with the predicted enrichment obtained by a material balance.
In chapter 5, photoluminescent GeO centers metabolically doped in
diatom biosilica reported antibody functionalization by up to a 50 times
enhancement in the PL from the baseline PL of the native biosilica without
germanium. The antibody functionalized biosilica PL signal was controlled
by reduction of the amine concentration, which tuned the PL signal to report
immunocomplex formation on the diatom biosilica surface. The germanium
doped biosilica reported immunocomplex formation with a PL signal which
was three times more intense than the PL signal of the native biosilica without
germanium upon immunocomplex formation.
Research Significance
Photoluminescent silica based nano- and micro- particles, films, and
meso structures are a desirable biosensor transducer platform because it is
biologically benign, non-toxic, water stable, has a large surface area, is a label
free detection method, and inherently exhibits a low signal background. [1-4]
However, industrial fabrication of photoluminescent silica is challenged.
Silica nano- and micro-particles must be doped with a luminophore upon
synthesis, or porous silicon must be ultrasonicated into smaller, typically non-
140
uniform pieces to yield photoluminescent particles. [2, 5-7] Films are made
photoluminescent by etching crystalline silicon wafers in hydrofluoric acid and
depending on the reaction conditions, the morphology and photoluminescence
can vary greatly. [8] Mesoporous silica is intrinsically photoluminescent,
however the synthesis conditions can significantly change the
photoluminescence [9] and often it is doped with dyes, tin, and other
luminescent materials to increase the PL intensity. [10, 11]
This is the first study to show that silica, fabricated by diatom cells,
which is intrinsically photoluminescent and contained within a massively
reproducible micro-structured form factor can be harnessed as a biosensor
platform and transducer to report biomolecules. Diatom biosilica is devoid of
current photoluminescent silica fabrication challenges, yet has all of the
desirable properties of photoluminescent silica.
There has been recent focus on doping silica with Ge, but not for
biosensor applications. Ge doped silica has a higher carrier mobility and
exhibits enhanced photoluminescence properties important for microelectronic
device applications such as the next generation of display devices. [12-14]
Current methods to fabricate Ge doped silica employ top down, low
throughput processes such as laser ablation, ion implantation, and cosputtering, which require extremes of temperature, pressure, and power. To
date, no studies have utilized the enhanced photoluminescence properties of Ge
doped silica for the purpose of biosensing.
Our research is the first to show that Ge doped biosilica can be
fabricated biologically using a diatom cell platform. Furthermore, biologically
fabricated Ge doped biosilica exhibits the same unique properties as
industrially fabricated Ge doped silica. Our research shows that PL based
141
sensing of diatom biosilica and Ge doped diatom biosilica can be combined to
yield a material with intense PL properties for sensing. Diatom biosilica has
nano- and micro-scale features which are contained within a single microscale
form factor, which is impossible to achieve with current fabrication methods.
Future Work
Integrating the PL based biosensing properties of Ge doped biosilica
frustules onto an optical platform for multi-plexed sensing was beyond the
scope of this study and is reserved for future work. It is envisioned that an
optical platform covered in frustules, each functionalized with a different
antibody, could report immunocomplex formation by enhanced
photoluminescence in a certain pattern. Furthermore, since the PL
enhancement was attributed to the attachment of nucleophilic ligands, this
same PL enhancement should be observed upon the attachment of living cells
or molecules with nucleophilic ligands. Contrary, a quenched PL signal would
likely be observed upon attachment with electrophilic ligands.
Comprehensive Conclusion
The photoluminescence of nanostructured diatom biosilica was
harnessed as a transducer to report immunocomplex formation by enhanced
photoluminescence. Metabolic doping of germanium into the nanostructured
diatom biosilica enhanced the intrinsic baseline photoluminescence intensity
which allowed for enhanced photoluminescence based immunocomplex
detection.
142
References
[1]
Wang, L.; Zhao, W.; Tan, W. Nano Res.2008, 1, 99-115.
[2]
Knopp, D.; Tang, D.; Niessner, R. Anal. Chim. Acta 2009, 647, 14-30.
[3]
Gu, Z.; Chen, X.-Y.; Shen, Q.-D.; Ge, H.-X.; Xu, H.-H. Polymer 2010,
51, 902-907.
[4]
Sailor, M.J.; Wu, E.C. Adv. Funct. Mater. 2009, 19, 3195-3208.
[5]
Santra, S.; Zhang, P.; Wang, K.; Tapec, R.; Tan, W. Anal. Chem. 2001,
73, 4988-4993.
[6]
Heinrich, J.L.; Curtis, C.L.; Credo, G.M.; Sailor, M.J.; Kavanagh, K.L.
Science 1992, 255, 66-68.
[7]
Bley, R.A.; Kauzlarich, S.M.; Davis, J.E.; Lee, H.W.H. Chem. Mater.
1996, 8, 1881-1888.
[8]
Zhang, X.G. J. Electrochem. Soc. 2004, 151, C69-C80.
[9]
Anedda, A.; Carbonaro, C.M.; Clemente, F.; Corpino, R.; Ricci, P.C.
Mater. Sci. Eng.,C 2003, 23, 1073-1076.
[12]
Gao, T.; Bao, X.M.; Yan, F.; Tong, S. Phys. Lett. A 1997, 232, 321325.
[13]
Shen, J.K.; Wu, X.L.; Bao, X.M.; Yuan, R.K.; Zou, J.P.; Tan C. Phys.
Lett. A 2000, 273, 208-211.
[14]
Li, J.; Wu, X.L.; Yang, Y.M.; Yang, X.; Bao, X.M. Phys. Lett. A 2003,
314, 299-303.
[10]
Liu, Z.C.; Chen, H.R.; Huang, W.M.; Gu, J.L.; Bu, W.B.; Hua, Z.L.;
Shi, J.L. Microporous Mesoporous Mater. 2006, 89, 270-275.
[11]
Sokolov, I.; Volkov, D.O. J. Mater. Chem. 2010, 20, 4247-4250.
143
BIBLIOGRAPHY
Anedda, A.; Carbonaro, C.M.; Clemente, F.; Corpino, R.; Ricci, P.C. Mater.
Sci. Eng.,C 2003, 23, 1073-1076.
Alessi, A.; Agnello, S.; Gelardi, F.M.; Grandi, S.; Magistris, A.; Boscaino, R.
Opt. Express 2008, 16, 7, 4895-4900.
Alessi, A.; Agnello, S.; Ouerdane, Y.; Gelardi, F.M. J. Phys.-Condens. Mat.
2011, 23, 1-6.
Arai, N.; Tsuji, H.; Nakatsuka, H.; Kojima, K.; Adachi, K.; Kotaki, H.;
Ishibashi, T.; Gotoh, Y.; Ishikawa, J. Mater. Sci. Eng., B 2008, 147, 230-234.
Azam, F.; Hemmingsen, B.B.; Volcani, B.E. Arch. Mikrobiol. 1973, 92, 11-20.
Barth, S.; Hernandez-Ramirez, F.; Holmes, J.D.; Romano-Rodriguez, A. Prog.
Mater. Sci. 2010, 55, 563-627.
Bekiari, V.; Lianos, P. J. Nanosci. Nanotechnol. 2006, 6, 372-376.
Bismuto, A.; Setaro, A.; Maddalena, P.; De Stefano, L.; De Stefano, M. Sens.
Actuators, B 2008, 130, 396-399.
The Biology of Diatoms, edited by D. Werner, Botanical Monographs,
University of California Press, Berkeley, CA (1977).
Blättler, T.; Huwiler, C.; Ochsner, M.; Städler, B.; Solak, H.; Vörös, J.;
Grandin, H.M. J. Nanosci. Nanotechnol. 2006, 6, 2237-2264.
Bley, R.A.; Kauzlarich, S.M.; Davis, J.E.; Lee, H.W.H. Chem. Mater. 1996, 8,
1881-1888.
Blin, J.L.; Carteret, C. J. Phys. Chem. C. 2007, 111, 14380-14388.
Bonanno, L.M.; DeLouise, L.A. Langmuir 2007, 23, 5817-5823.
144
Borchers, K.; Weber, A.; Brunner, H.; Tovar, G.E.M. Anal. Bioanal. Chem.
2005, 383, 738-746.
Briand, E.; Salmain, M.; Compère, C.; Pradier, C.-M. Biosens. Bioelectron.
2007, 22, 2884-2890.
Brzezinski, M.A.; Conley, D.J. J. Phycol. 1994, 30, 45-55.
Brzezinski, M.A.; Olson, R.J.; Chisholm, S.W. Mar. Ecol. Prog. Ser. 1990, 67,
83-96.
Butcher, K.S.A.; Ferris, J.M.; Phillips, M.R.; Fouquet-Wintrebert, M.; Jong
Wah, J.W.; Jovanovic, N.; Vyverman, W.; Chepurnov, V.A. Mater. Sci. Eng.,
C 2005, 25, 658-663.
Cai, Y.; Dickerson, M.B.; Haluska, M.S.; Kang, Z.; Summers, C.J.; Sandhage,
K.H. J. Am.Ceram. Soc. 2007, 90, 1304-1308;
Carbonaro, C.M.; Clemente, F.; Corpino, R.; Ricci, P.C.; Anedda, A. J. Phys.
Chem. B 2005, 109, 14441-1444.
Calvert, P. Chem. Mater. 2001, 13, 3299-3305.
Carlos, L. D.; Sa´ Ferreira, R. A.; Pereira, R. N.; Assunca˜o, M.; de Zea
Bermudez, V. J. Phys. Chem. B 2004, 108, 14924.
Céreyon, A.; Champagnon, B.; Martinez, V.; Maksimov, L.; Yanush, O.;
Bogdanov, V.N. Opt. Mater. 2006, 28, 1301-1304.
Chaniotakis, N.; Sofikiti, N. Anal. Chim. Acta 2008, 615, 1-9.
Chiappino, M.L.; Azam, F.; Volcani, B.E. Protoplasma 1977, 93, 191-204.
Choi, W.K.; Ho, Y.W.; Ng, S.P.; Ng, V. J. Appl. Phys. 2001, 89, 2168-2172;
Christman, K.L.; Enriquez-Rios, V.D.; Maynard, H.D. Soft Matter 2006, 2,
928-939.
Cojocaru, C.V.; F. Ratto, C. Harnagea, A. Pignolet, F. Rosei, Microelectron.
Eng. 2005, 80, 448-456;
145
Conway, H.L.; Harrison P.J. Mar. Biol. 1977, 43, 33-43.
Conway, H.L.; Harrison, P.J.; Davis, C.O. Mar. Biol. 1976, 35, 187-199.
Costacurta, S.; Malfatti, L.; Kidchob,T.; Takahashi, M.; Mattei, G.; Bello,V.;
Maurizio, C.; Innocenzi, P. Chem. Mater. 2008, 20, 3259-3265
Crawford, S.A.; Higgins, M.J.; Mulvaney, P.; Wetherbee, R. J. Phycol. 2001,
37, 543-554.
Dannhauser, T.; O’Neil, M.; Johansson, K.; Whitten, D.; McLendon, G. J.
Phys.Chem. 1986, 90, 6074-6076.
De Stefano, L.; Lamberti, A.; Rotiroti, L.; De Stefano, M. Acta Biomater.
2008, 4, 126-130.
De Stefano, L.; Rendina, I.; De Stefano, M.; Bismuto, A.; Maddalena, P. Appl.
Phys. Lett. 2005, 87, 233902.
De Stefano, L.; Lamberti, A.; Rotiroti, L.; De Stefano, M. Acta Biomater.
2008, 4, 126-130.
De Stefano, L.; Rotiroti, L.; De Stefano, M.; Lamberti, A.; Lettieri, S.; Setaro,
A.; Maddalena, P. Biosens. Bioelectron., 2009, 24, 1580-1584.
Deng, G.; Markowitz, M.A.; Kust, P.R.; Gaber, B.P. Mat. Sci. Eng. C 2000,
11, 165-172.
Di Francia, G.; LaFerrara, V.; Manzo, S.; Chiavarini, S. Biosens. Bioelectron.
2005, 21, 661-665.
Dorfman, A.; Kumar, N.; Hahm, J. Adv. Mater. 2006, 18, 2685-2690.
Dorfman, A.; Kumar, N.; Hahm, J., Langmuir 2006, 22, 4890.
Dudley, S.; Kalem, T.; Akinc, M. J. Am. Ceram. Soc 2006, 89, 2434-2439.
Dujardin, E.; Mann, S. Adv. Mater. 2002, 14,11, 775-788.
Durben, D.J.; Wolf, G.H. Phys. Rev. B. 1991, 43, 2355-2363.
146
Dybiec, M.; Chornokur, G.; Ostapenko, S.; Wolcott, A.; Zhang, J. Z.; Zajac,
A.; Phelan, C.; Sellers, T.; Gerion, D. Appl. Phys. Lett. 2007, 90, 263112.
Erickson, D.; Mandal, A.; Yang, H.J.; Cordovez, B. Microfluid. Nanofluid.
2008, 4, 33-52.
Fan, H.J.; Werner, P.; Zacharias, M. Small 2006, 2, 700-717.
Gale, D.K.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. Adv. Funct. Mater.
2009, 19, 926-933.
Gale, D.K.; Jeffryes, C.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. J.
Mater. Chem. 2011, 21, 10658-10665.
Gale, D.K.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. Germanium
Enrichment in the Girdle Band of the Centric Diatom Cyclotella sp., 2011,
manuscript in preparation.
Galeener, F.L.; Mikkelsen, J.C.; Geils, R.H.; Mosby, W.J. Appl. Phys. Lett.
1978, 32, 34-36.
Gao, T.; Bao, X.M.; Yan, F.; Tong, S. Phys. Lett. A, 1997, 232, 321–325.
Garcia, M.A.; Paje, S.E.; Villagas, M.A.; Llopis, J. Mater. Lett. 2000, 43, 2326.
Gélabert, A.; Pokrovsky, O.S.; Schott, J.; Boudou, A.; Feurtet-Mazel, A.;
Mielczarski, J.;
Mielczarski, E.; Mesmer-Dudons, N.; Spalla, O. Geochim. Cosmochim. Acta
2004, 68, 4039-4058.
Gendron-Badou, A.; Coradin, T.; Maquet, J.; Fröhlich, F.; Livage, J. J. NonCryst. Solids 2003, 316, 331-337.
Gish, D.A.; Nsiah, F.; McDermott, M.T.; Brett, M.J. Anal. Chem. 2007, 79,
4228-4232.
Glinka, Y.D.; Zyubin, A.S.; Mebel, A.M.; Lin, S.H.; Hwang, L.P.; Chen, Y.T.
Chem. Phys. Lett. 2002, 358, 180-186;
147
Glinka, L.D.; Lin, S.-H.; Chen, Y.T. Phys. Rev. B: Condes. Matter. 2002, 66,
035404-1-10.
Gooding, J.J.; Wasiowych, C.; Barnett, D.; Hibbert, D.B.; Barisci, J.N.;
Wallace, G.G. Biosens. Bioelectron. 2004, 20, 260-268.
Grandi, S.; Mustarelli, P.; Agnello, S.; Cannas, M.; Cannizzo, A. J. Sol-Gel
Sci. Techn. 2003, 26, 915-918.
Grassman, T.J.; Bishop, S.R.; Kummel, A.C. Surf. Sci. 2008, 602, 2373-2381.
Gu, Z.; Chen, X.-Y.; Shen, Q.-D.; Ge, H.-X.; Xu, H.-H. Polymer 2010, 51,
902-907.
Gu, Z.; Liu, F.; Howe, J.Y.; Paranthaman, M.P.; Pan, Z. Crys. Growth
Des.2009, 9, 35-39.
Gulino, A.; Giuffrida, S.; Mineo, P.; Purrazzo, M.; Scamporrino, E.;
Ventimiglia, G.; van der Boom, M.E.; Fragalá, I. J. Phys. Chem. B 2006, 110,
16781-16786.
Gutu, T.; Gale, D.K.; Jeffryes, C.; Wang, W.; Chang, C.-H.; Rorrer, G.L.; Jiao,
J. J. Nanomater. 2009, 860536.
Harrison, P.J.; Waters, R.E.; Taylor, F.J.R. J. Phycol. 1980, 16, 25-28.
He, N.; Ge, S.; Yang, C.; Hu, C., Gu, M. Mater. Res. Bull.2004, 39, 19311937.
He, N.Y.; Ge, S.X.; Yang, C.; Cao, J.M.; Gu, M. Mater. Lett., 2004, 58, 3304–
3307.
Heinrich, J.L.; Curtis, C.L.; Credo, G.M.; Sailor, M.J.; Kavanagh, K.L. Science
1992, 255, 66-68.
G.T. Hermanson, Bioconjugate Techniques, Academic Press, London, 1996.
Hildebrand, M.; York, E.; Kelz, J.I.; Davis, A.K.; Frigeri, L.G.; Allison, D.P.;
Doktycz, M.J. J. Mater. Res. 2006, 21, 2689-2698.
Hildebrand, M.; Frigeri, L.G.; Davis, A.K. J. Phycol. 2007, 43, 730-740.
148
Hu, J.; Jiang, Y.; Meng, X.; Lee, C.-S.; Lee, S.-T. Small, 2005, 1, 429-438.
Huang, J.-G; Lee, C.-L.; Lin, H.-M.; Chuang, T.-L.; Wang, W.-S.; Juang, R.H.; Wang, C.-H.; Lee, C.-K.; Lin, S.-M.; Lin, C.-W. Biosens. Bioelectron.
2006, 22, 519-525.
Humbert, B. J. Non-Cryst. Solids 1995, 191, 29-37.
Ibach, H.; Rowe, J.E. Phys. Rev. B 1974, 10, 710-718.
Inaki,Y.; Yoshida, H.; Yoshida, T.; Hattori, T. J. Phys. Chem. B 2002, 106,
9098-9106.
Jane, A.; Dronov, R.; Hodges, A.; Voelcker, N.H. Trends Biotechnol. 2009,
27, 230-239.
Jeffryes, C.; Gutu, T.; Jiao, J.; Rorrer, G.L. Mater. Sci. Eng. C 2008, 28, 107118.
Jeffryes, C; Solanki, R.; Rangineni, Y.; Wang, W.; Chang, C.-H.; Rorrer, G.L.
Adv. Mater. 2008, 20, 2633-2637.
Jeffryes, C.; Gutu, T.; Jiao, J.; Rorrer, G.L. ACS Nano 2008, 2, 2103-2112.
Jia, R.P.; Shen, Y.; Luo, H.Q.; Chen, X.G.; Hu, Z.D.; Xue, D.S. Thin Solid
Films 2005, 471, 264-269.
Jianrong, C.; Yuqing, M.; Nongyue, H.; Xiaohua, W.; Sijiao, L. Biotechnol.
Adv. 2004, 22, 505-518.
Kaczmarska, I.; Beaton, M.; Benoit, A.C.; Medlin, L.K. J. Phycol. 2005, 42,
121-138.
Kammer, M.; Hedrich, R.; Ehrlich, H.; Popp, J.; Brunner, E.; Krafft, C. Anal.
Bioanal. Chem. 2010, 398, 509-517.
Kanjilal, A; Rebohle, L.; Voelskow, M.; Skorupa, W.; Helm, M. Appl. Phys.
Lett. 2009, 94, 051903.
Knopp, D.; Tang, D.; Niessner, R. Anal. Chim. Acta 2009, 647, 14-30.
149
Ko, M.C.; Meyer, G.J. Chem. Mater. 1996, 8, 2686-2692.
Kovalchuk, T.; Sfihi, H.; Kostenko, L.; Zaitsev, V.; Fraissard, J. J. Colloid
Interface Sci. 2006, 302, 214-229.
Kröger, N.; Poulsen, N. Annu. Rev. Genet. 2008, 42, 83-107.
Kroger, N.; Deutzmann, R.; Sumper, M. Science 1999, 286, 5 1129-1132.
Lauerhaas, J.M.; Sailor, M.J. Science 1993, 261, 1567-1568.
Lee, D.-H.; Gutu, T.; Jeffryes, C.; Rorrer, G.L.; Jiao, J.; Chang,C.-H.
Electrochem. Solid-State Lett. 2007, 10, K13-K16.0.2;
Lee, C.-H.; Wang, W.; Gutu, T.; Jeffryes, C.; Rorrer, G.L.; Jiao, J.; Chang, C.H. J. Mater. Chem. 2008, 18, 3633-3635;
Lettieri, S.; Setaro, A.; De Stefano, L.; De Stefano, M.; Maddalena, P. Adv.
Funct. Mater. 2008, 18, 1257-1264.
Lewin, J. Phycologia 1966, 6, 1-12.
Lewis, S.E.; DeBoer, J.R.; Gole, J.L.; Hesketh, P.J. Sens. Actuators, B 2005,
110, 54-56.
Lewis, S.E.; DeBoer, J.R.; Gole, J.L.; Hesketh, P.J. Sens. Actuators, B 2005,
110, 54-56.
Li, J.; Wu, X.L.; Yang, Y.M.; Yang, X.; Bao, X.M. Phys. Lett. A 2003, 314,
299-303.
Lin, K.-C.; Kunduru, V.; Bothara, M.; Rege, K.; Prasad, S.; Ramakrishna, B.L.
2010, 25, 2336-2342.
Liu, T.-Y.; Liao, H.-C.; Lin, C.-C.; Hu, S.-H.; Chen, S.-Y. Langmuir 2006, 22,
5804.
Liu, Z.C.; Chen, H.R.; Huang, W.M.; Gu, J.L.; Bu, W.B.; Hua, Z.L.; Shi, J.L.
Microporous Mesoporous Mater. 2006, 89, 270-275.
150
Lockwood, D.J.; J. Mater. Sci.: Mater. Electron. 2009, 20, S235–S244.
Losic, D.; Rosengarten, G.; Mitchell, J.G.; Voelcker, N.H. J. Nanosci.
Nanotechnol. 2006, 6, 982-989.
Losic, D.; Triani, G.; Evans, P.J.; Atanacio, A.; Mitchell, J.G.; Voelcker, N.H.
J. Mater. Chem. 2006, 16, 4029-4034.
Losic, D.; Mitchell, J.G.; Voelcker, N.H. Adv. Mater. 2009, 21, 2947-2958.
McCreery, R.L. Raman Spectroscopy for Chemical Analysis; WileyInterscience: New York, 2000.
Marazuela, M.D.; Moreno-Bondi, M.C. Anal. Bioanal. Chem. 2002, 372-664682.
Martin-Jézéquel, V.; Hildebrand, M.; Brzezinski, M.A. J. Phycol. 2000, 36,
821-840.
Mendes, P.M.; Yeung, C.L.; Preece, J.A. Nanoscale Res. Lett. 2007, 2, 373384.
Messina, F.; Agnello, S.; Boscaino, R.; Cannas, M.; Grandi, S.; Quartarone, E.
J. Non-Crys. Solids 2007, 353, 670-673.
Micoulaut, M.; Cormier, L.; Henderson, G.S. J. Phys.-Condens. Mat. 2006, 18,
R753-R784.
Molle, A.; Bhuiyan, M.N.K.; Tallarida, G.; Fanciulli, M. Mater. Sci. Semicond.
Process. 2006, 9, 673–678.
Morse, D.E. Trends Biotechnol. 1999, 17, 230-232.
Nassif, N.; Livage, J. Chem. Soc. Rev. 2011, 40, 849-859.
Neustrev, V.B. J. Phys. Condens. Matt. 1994, 6, 6901-6936.
Nisonoff, A.; Hopper, J.E.; Spring, S.B The Antibody Molecule, Academic
Press, New York, 1975.
Oh, J.; Campbell, J.C. J. Electron. Mater. 2004, 33, 364-367.
151
Oswald, S.; Schmidt, B.; Heinig, K.-H. Surf. Interface Anal. 2000, 29, 249254;
Paul, S.; Nishiyama, Y.; Planque, S.; Karle, S.; Taguchi, H.; Hanson, C.;
Weksler, M.E. Springer Semin. Immun. 2005, 26, 485-503.
Pelissier, B.; Kambara, H.; Godot, E.; Veran, E.; Loup, V.; Joubert, O.
Microelectron. Eng. 2008, 85, 151-155.
Peng, L.; Qisui, W.; Xi, L.; Chaocan, Z. Colloids Surf., A 2009, 334, 112-115.
Prabhakaran, K.; Maeda, F.; Watanabe, Y.; Ogina, T. Appl. Phys. Lett. 2000,
76, 2244-2246;
Qin, T.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. J. Nanosci. Nanotechnol.
2008, 8, 1-7.
Qin, T.; Gutu, T.; Jiao, J.; Chang, C.-H.; Rorrer, G.L. ACS Nano 2008, 2,
1296-1304.
Rebohle, L.; von Borany, J.; Grötzschel, R.; Markwitz, A.; Schmidt, B.;
Tyschenko, I.E.; Skorupa, W.; Fröb, H.; Leo, K. Phys. Status Solidi A, 1998,
165, 31–35.
Riegel, B.; Hartmann, I.; Kiefer,W.; Groß, J.; Fricke, J. J. Non-Cryst. Solids
1997, 211, 294-298.
Rorrer, G.L.; Chang, C.-H.; Liu, S.-H.; Jeffryes, C.; Jiao, J.; Hedberg, J.A. J.
Nanosci. Nanotechnol. 2005, 5, 41-49.
Round, F.E. Nova Hedwagia 1972, 23:449-63.
Rückschloss, M.; Wirschem. Th.; Tamura, H.; Ruhl, G.; Oswald, J.; Veprek, S.
J. Lumin. 1995, 63, 279-287.
Rusmini, F.; Zhong, Z.; Feijen, J. Biomacromolecules 2007, 8, 1775-1789
Sailor, M.J.; Wu, E.C. Adv. Funct. Mater. 2009, 19, 3195-3208.
152
Santra, S.; Zhang, P.; Wang, K.; Tapec, R.; Tan, W. Anal. Chem. 2001, 73,
4988-4993.
Sarangi, S.N.; Goswami, K.; Sahu, S.N. Biosens. Bioelectron. 2007, 22, 3086.
Sarikaya, M.; Tamerler, C.; Schwarz, D.T.; Baneyx, F. Annu. Rev. Mater. Res.
2004, 34, 373-408.
Scala, S.; Bowler, C. Cell. Mol. Life Sci. 2001, 58-1666-1673.
Schaferling, M.; Nagl, S. Anal. Bioanal. Chem. 2006, 385, 500-517.
Schechter, I.; Ben-Chorin, M.; Kux, A. Anal. Chem. 1995, 67, 3727-3732.
Setaro, A.; Lettieri, S.; Maddalena, P.; De Stefano, L. Appl. Phys. Lett. 2007,
91, 051921.
Shang, Q.; Wang, H.; Yu, H.; Shan, G.; Yan, R. Colloid Surf. A 2007, 294, 86.
Shieh, J.M.; Cho, A.-T.; Lai, Y.-F.; Dai, B.-T.; Pan, F.-M.; Chao, K.-J.
Electrochem. Solid-State Lett. 2004, 7, G319-G322.
Shen, J.K.; Wu, X.L. Bao, X.M.; Yuan, R.K.; Zou, J.P.; Tan, C. Phys Lett. A
2000, 273, 208-211.
Sokolov, I.; Volkov, D.O. J. Mater. Chem. 2010, 20, 4247-4250.
Starodub, N.F.; Fedorenko, L.L.; Starodub, V.M.; Dikij, S.P.; Svechnikov,
S.V. Sens. Actuators, B 1996, 35-36, 44-47.
Stewart, M.P; Buriak, J.M. Adv. Mater. 2000, 12, 859.
Sumper, M.; Kroger, N. J. Mater. Chem. 2004, 14, 20-59-2065.
Tamura, H.; R c schloss, M.; Wirschem, T.; Vepře , S. Appl. Phys. Lett.
1994, 65, 1537-1539;
Tesson, B.; Hildebrand, M. J. Struct. Biol. 2010, 169, 62-74.
Thamatrakoln, K.; Hildebrand, M. Plant Physiol. 2008, 146, 1397-1407.
153
Thomas, W.H.; Dodson, A.N. Mar. Biol. 1974, 27, 11-19.
Townley, H.E.; Woon, K.L.; Payne, F.P.; White-Cooper, H.; Parker, A.R.
Nanotechnology 2007, 18, 295101.
Townley, H.E.; Parker, A.R.; White-Cooper, H. Adv. Funct. Mater. 2008, 18,
369-374.
Tsybeskov, L.; Vandyshev, J.V.; Fauchet, P.M. Phys. Rev. B. Condes. Matter
1994, 49, 7821-7824.
Tuttle, P.V.; Rundell, A.E.; Webster, T.J. Int. J. Nanomed. 2006, 1, 497-505.
Vaccaro, L.; Cannas, M.; Boscaino, R. Solid State Commun. 2008, 146, 14151.
Vo-Dinh, T.; Cullum, B. J. Anal. Chem. 2000, 366, 540-551.
Wang, C.; Yap, F.L.; Zhang, Y. Colloid Surface B 2005, 46, 255-260.
Wang, J.-H.; Wang, H.-Q.; Zhang, H.-L.; Li, X.-Q.; Hua, X.-F.; Huang, Z.-L.;
Zhao, Y.-D. Colloid Surf. A 2007, 305, 48.
Wang, L.; Zhao, W.; Tan, W. Nano Res.2008, 1, 99-115.
Wang, Q.; Kuo, Y.; Wang, Y.; Shin, G.; Ruengruglikit, C.; Huang, Q. J.
Phys.Chem. B. 2006, 110, 16860.
Wang, W.; Gutu, T.; Gale, D.K.; Jiao, J.; Rorrer, G.L.; Chang, C.-H. J. Am.
Chem. Soc. 2009, 131, 4178-4179.
Weatherspoon, M.R.; Allan, S.M.; Hunt, E.; Cai, Y.; Sandhage, K.H. Chem.
Commun. 2005, 651-653.
Westcott, S.L.; Oldenberg, S.J.; Lee, T.R.; Halas, N.J. Langmuir 1998, 14,
5396-5401.
White, S.N. Chem. Geol. 2009, 259, 240-252.
Wopenka, B.; Pasteris, J.D. Anal. Chem. 1987, 59, 2165-2170.
154
Wu, J.; Coffer, J.L. Chem. Mater. 2007, 19, 6266-6276.
Wu, J.; Coffer, J.L.; Wang, Y.; Schulze, R. J. Phys. Chem. C 2009, 113, 12-16.
Wu, J.; Punchaipetch, P.; Wallace, R.M.; Coffer, J.L. Adv. Mater. 2004, 16,
1444-1448.
Wu, L.-Q.; Payne, G.F. Trends Biotechnol. 2004, 22, 593-599.
Wu, X.M.; Lu, M.J.; Yao, W.G. Surf. Coat. Technol. 2002, 161, 92-95;
Wu, X.L.; Gao, T.; Bao, X.M.; Yan, F.; Jiang, S.S.; Feng, D. J. Appl. Phys.
1997, 82, 2704-2706;
Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim, F.;
Yan, H. Adv. Mater. 2003, 15, 353-389.
Xu, Z.W.; Ngan, A.H.W.; Hua, W.Y.; Meng, X.K. Appl. Phys. A: Mater. Sci.
Process., 2005, 81, 459–463.
Yang, Y.; Addai-Mensah, J.; Losic, D. Langmuir 2010, 26, 14068-14072.
Yang, W.; Lopez, P.J.; Rosengarten, G.R. Analyst 2011, 136, 42-53.
Yao, Z.; Yang, M. Sens. Actuators, B 2006, 117, 93-98.
Yuan, P.; He, H.P.; Wu,D.Q.; Wang, D.Q.; Chen, L.J. Spectrochim. Acta. A
2004, 60, 2941-2945.
Yuan, P.; Wu, D.Q.; He, H.P.; Lin, Z.Y. Appl. Surf. Sci 2004, 227, 30-39.
Zacharias, M.; Fauchet, P.M. Appl. Phys. Lett. 1997, 71, 380-382;
Zatsepin, A.F.; Fitting, H.-J.; Kortov, V.S.; Pustovarov, V.A.; Schmidt, B.;
Buntov, E.A. J. Non-Cryst. Solids 2009, 355, 61-67.
Zhang, S. Nat. Biotechnol. 2003, 21, 10, 1171-1178.
Zhang, Q.; Shin, Y.J.; Hua, F.; Saraf, L.V.; Matson, D.W. J. Nanosci.
Nantechnol. 2008, 8, 3008-3012.
155
Zhang, X.G. J. Electrochem. Soc. 2004, 151, C69-C80.
Zhu, H.; Han, Y.; Wehrspohn, R.B.; Godet, C.; Etemadi, R.; Ballutaud, D. J.
Appl. Phys. 1998, 83, 5386-5393;
Zhu, Y.; Yuan, C.L.; Ong, P.P. J. Appl. Phys. 2003, 93, 6029-6033.
Zhu, Y.; Yuan, C.L.; Quek, S.L.; Chan, S.S.; Ong, P.P. J. Appl. Phys. 2001,
90, 5318-5321.
Zhuravlev, L.T. Colloids Surf., A 2000, 173, 1-38;
Zurzola, C.; Bowler, C. Plant Physiol. 2001, 127, 1339-1345.
156
APPENDIX
157
Procedure A-1.
Amine Functionalization of H2O2 Treated Diatom Frustules
With 3-aminopropyltrimethoxysilane (1X APS Dilution, 0.143 mol
APS/L)
Debra Gale (revised December 20, 2010)
Specialized Equipment
 Reactitherm
 Reactitherm 3.0 mL reactivial
 Reactivial conical stir bar to fit reactivial
 6 mL amber glass vial
Specialized Equipment
 10 mg of diatom biosilica which has been treated with hydrogen peroxide
according to procedure
Reagents
 3-aminopropyltrimethoxysilane (APS) (Sigma Aldrich 09326-100ML),
H2N(CH2)3Si(OCH3)3, 179.29 g/mol, density 1.01 g/mL
 Anhydrous Reagent grade ethanol, C2H2O, 46.06 g/mol
Procedure
1. Complete all Procedure steps in a laminar flow hood wearing eyewear and
gloves.
2. Suspend 10 mg of hydrogen peroxide treated diatom frustules in a 2.0 mL
of AR-grade ethanol in a 3.0 mL reactivial with a reactivial stir bar.
3. Add to the reactivial 50 µL of 3-aminopropyltrimethoxysilane (APS).
4. Place reactivial in reactitherm in laminar flow hood and turn heat on “high”
at setting 1.75 and turn on stirring mechanism. This setting is calibrated to
80ºC.
5. Allow frustule and APS mixture to stir at 80ºC for 2 hours, periodically
double checking the temperature to make sure it is still 80°C.
6. After 2 hours, turn off the heat on the reactitherm, but do not remove the
reactivial from the reactitherm.
7. Allow the frustule and APS mixture to stir for another 22 hours without
any heating.
8. After the 24 hour reaction time, wash the biosilica with ethanol by
centrifugation of the biosilica and removal and replacement of the ethanol.
158
9. If the aminated biosilica is being deposited on cover slips for future
reactions, then anneal the biosilica thin film at 90°C for 1 hour in a preheated furnace before continuation with functionalization steps.
** In hand calculations
0.143 mol APS/L
0.0286 mol APS/g SiO2
1.83 mol APS/mol SiO2
1.588x10-13 mol APS/frustule
9.568x1010 APS molecules/frustule
159
Procedure A-2.
Amine Functionalization of H2O2 Treated Diatom Frustules
With 3-aminopropyltrimethoxysilane (5000x APS Functionalization)
Debra Gale (revised December 20, 2010)
Specialized Equipment
 Reacti-therm
 Reactitherm 3.0 mL reactivial
 Reactivial conical stirbar to fit reactivial
Specialized Equipment
 10 mg of diatom biosilica which has been treated with hydrogen peroxide
according to procedure

Reagents
 3-aminopropyl trimethoxysilane (APS) (Sigma Aldrich 09326-100ML),
H2N(CH2)3Si(OCH3)3, 179.29 g/mol
 Anhydrous Reagent grade ethanol, C2H2O, 46.06 g/mol
Procedure
10. Complete all Procedure steps in a laminar flow hood wearing eyewear and
gloves.
11. Suspend 10 mg (or less if not enough material is available) of hydrogen
peroxide treated diatom frustules in a 1.0 mL of AR-grade ethanol in a 3.0
mL reactivial with a reactivial stir bar.
12. Prepare APS Superstock 1: 0.05 mL APS in 100 mL of ethanol
13. Prepare APS Superstock 2 (final solution): 1.0 mL SS#1 in 50 mL EtOH
14. Add 1 mL of Superstock 2 to the reactivial containing 1 mL of ethanol and
10 mg of frustules.
15. Place reactivial in reactitherm in laminar flow hood and turn heat on “high”
at setting 1.75 and turn on stirring mechanism. This setting is calibration
to 80ºC.
16. Allow frustule and APS mixture to stir at 80ºC for 2 hours, periodically
double checking the temperature.
17. After 2 hours, turn off the heating do not remove the reactivial from the
reactitherm.
18. Allow the frustule and APS mixture to stir for another 22 hours without
any heating.
160
19. After the 24 hour reaction time, wash the biosilica with ethanol and by
centrifugation and pipetting off the ethanol.
20. If the aminated biosilica is being deposited on cover slips for future
reactions, then anneal at 90°C for 1 hour before continuing with
functionalization steps.
** In hand calculations
2.864x10-5 mol APS/L
5.72x10-6 mol APS/g SiO2
3.665x10-4 mol APS/mol SiO2
3.177x10-17 mol APS/frustule
1.9136x107 APS molecules/frustule
161
Procedure A-3.
Antibody Rabbit Immunoglobulin G (IgG) Crosslinking of
Bis[sulfosuccinimidyl]suberate (BS3), 3-aminopropyltrimethoxysilane
(APS) Functionalized Diatom Frustule Films on Round Coverslips
Debra Gale (revised August 25, 2009)
Specialized Equipment
 6 Well Polystyrene Plate
 VWR shaker, 50 rpm
 1.5 mL microcentrifuge tube
Reagents
 Phosphate buffered saline (PBS) made the procedure Phosphate
Buffered Saline Preparation, pH 7.6, 0.01M Phosphate, 0.15 M NaCl
08-25-09
 ImmunoPure Rabbit IgG (AB), Whole Molecule (Pierce Biotechnology
# 31235),
MW = 150,000 g/mol, 11.7 mg/mL
Sample
 3-aminopropyltrimethoxysilane (APS), Bis[sulfosuccinimidyl]suberate
(BS3) Functionalized Diatom Frustule film on a round coverglass slip,
0.06 mg H2O2 treated frustules in a 5 mm diameter film, see Procedure
Diatom Frustule Film Preparation
1. Add 1.8 mL of PBS to one well of the 6 well plate.
2. Combine 0.02 mL ImmunoPure Rabbit IgG with 9.98 mL PBS in a
microcentrifuge tube. This will yield a solution of the mass
concentration 0.0234 mg AB/mL Rabbit IgG.
3. Add 0.20 mL of the 0.0234 mg/mL Rabbit IgG to the 1.8 mL PBS
filled well of the 6 well plate. This will yield a mass concentration of
rabbit IgG in the well of 0.00234 mg Rabbit IgG/mL.
4. Place the APS-BS3 functionalized diatom film covered coverslip film
side down on the PBS, Rabbit IgG solution. The surface tension of the
PBS, IgG mixture will prevent the coverslip from sinking.
5. Place the 6 well plate on the shaker at 50 rpm for 2 hours.
6. Using tweezers carefully remove the coverslip.
162
7. Dip the coverslip in PBS to remove excess antibody. However, if
continuing to the antibody-antigen reaction step, then do not rinse and
place in well with antigen solution. If the backside of the coverslip is
wet from PBS, then it will sink to the bottom of the well during the
antigen functionalization step.
8. Place the coverslip film side up on a piece of paper for drying or
proceed to next reaction step.
163
Procedure A-4.
Antibody Rabbit Immunoglobulin G (IgG) Crosslinking of
Bis[sulfosuccinimidyl]suberate (BS3), 3-aminopropyltrimethoxysilane
(APS) Functionalized Diatom Frustules
Debra Gale (revised May 4, 2008)
Specialized Equipment
 Eppendorf 5414 Microcentrifuge
 2.0 mL microcentrifuge tube
 Barnstead Thermolyne Lab Quake Rotisserie, 8.0 rpm, Model # 415110
 Sterile Pipette tips which fit a 20-200 μL pipette
Reagents
 Phosphate buffered saline (PBS) made the procedure Phosphate
Buffered Saline Preparation, pH 7.6, 0.01M Phosphate, 0.15 M NaCl
08-25-09
 ImmunoPure Rabbit IgG, Whole Molecule (Pierce Biotechnology #
31235),
MW = 150,000 g/mol, 11.7 mg/mL
Procedure
1. Add 0.07 mL ImmunoPure Rabbit IgG to 1 mL (10 mg/mL) APS- BS3
functionalized diatom frustules in PBS in a microcentrifuge tube using
a pipette with a sterile tip.
2. Rotate for 2 hours on the Lab Quake Rotisserie at 8 rotations per
minute.
3. Centrifuge at 1,000 rpm (82 g) for 10 minutes.
4. Pipette off supernatant.
5. Replace supernatant with 1.0 mL PBS solution.
6. Repeat steps 3-5, twice to remove excel antibody.
164
Procedure A-5.
Antigen Rabbit Anti-IgG Crosslinking of 3-aminopropyltrimethoxysilane
(APS), Bis[sulfosuccinimidyl]suberate (BS3), Rabbit IgG Functionalized
Diatom Frustule Films on Round Coverslips
Debra Gale (revised August 25, 2009)
Specialized Equipment
 1.5 mL microcentrifuge tube
 Leica DMIL microscope
 GFP filter for Leica DMIL microscope (excitation: 470/40 nm,
emission: 525/50 nm)
 6 well polystyrene plate
 VWR shaker, 50 rpm
Reagents
 Phosphate buffered saline made the procedure Phosphate Buffered
Saline Preparation, pH 7.6, 0.01M Phosphate, 0.15 M NaCl 08-25-09
 ImmunoPure Goat Anti-Rabbit IgG, (H+L), MW = 150,000 g/mol, 2.4
mg/mL (Pierce # 31210)
 ImmunoPure Fluorescein Conjugated Goat Anti-Rabbit IgG (H+L)
(Pierce # 31635), 2 mg
Sample
 3-aminopropyltrimethoxysilane (APS), Bis[sulfosuccinimidyl]suberate
(BS3) , ImmunoPure Rabbit IgG Functionalized Diatom Frustule film
on a round coverglass slip, 0.060 mg H2O2 treated frustules as a 5 mm
diameter film
Antigen: ImmunoPure Goat Anti-Rabbit IgG, (H+L), MW = 150,000 g/mol,
2.4 mg/mL (Pierce # 31210)
The following procedure is for an anti-Rabbit IgG mass concentration of
0.0797 mg/mL anti-IgG, however the all of the following Anti-Rabbit IgG
mass concentrations were used: 0.0000797 mg/mL, 0.000797 mg/mL,
165
0.00797 mg/mL, 0.03 mg/mL, 0.0419 mg/mL, 0.06 mg/mL, 0.0797 mg/mL,
0.08395 mg/mL, 0.115 mg/ mL, 0.12 mg/ mL, 0.46 mg/mL, 2.3 mg/mL.
1. Add 1.8 mL of PBS to one well of the 6 well plate
2. Combine 0.4649 mL of 2.3 mg/mL ImmunoPure Goat Anti-Rabbit IgG
with 0.9351 mL of PBS in a microcentrifuge tube. This will yield a
solution of the mass concentration 0.0797 mg/mL anti-Rabbit IgG.
3. Add 0.20 mL of the 0.0797 mg/mL anti-Rabbit IgG solution to the PBS
filled well.
4. Place the APS, BS3, Rabbit IgG functionalized diatom frustules film
covered coverslip film side down on the PBS, anti-Rabbit IgG solution.
The surface tension of the PBS, anti-IgG mixture will prevent the
coverslip from sinking.
5. Place the 6 well plate on the shaker at 50 rpm for 2 hours.
6. Using tweezers carefully remove the coverslip.
7. Dip the coverslip in PBS to remove excess antibody.
8. Place the coverslip film side up on a piece of paper for drying.
Antigen: ImmunoPure Fluorescein (FITC) Conjugated Goat Anti-Rabbit IgG
(H+L), 2.0 mg, Pierce Product # 31635, excitation: 495, emission: 520, MW
151,207
The following procedure is for an anti-Rabbit IgG mass concentration of
0.0797 mg/mL anti-IgG.
1. Add 1.84 mL of PBS to one well of a six well plate.
2. Add 0.5 mg of FITC Conjugated Goat Anti-Rabbit IgG to 0.5 mL PBS
in a black microcentrifuge tube.
3. Invert the black microcentrifuge tube to mix the solution.
4. Add 0.1594 mL of the 1 mg/mL FITC Conjugated Anti-Rabbit IgG to
the PBS filled well.
5. Place the APS,BS3, Rabbit IgG functionalized diatom frustules film
coverslip film side down on the PBS, FITC-anti-Rabbit IgG solution.
The surface tension of the PBS, anti-IgG mixture will prevent the
coverslip from sinking.
6. Cover the 6 well plate in foil to protect the FITC from photobleaching.
7. Place the 6 well plate on the shaker at 60 rpm for 2 hours.
8. Using tweezers carefully remove the coverslip.
9. Dip the coverslip in PBS to remove excess antigen
166
10. Place the coverslip film side up on a piece of paper for drying in a dark
location.
11. Using the DMIL microscope equipped with a GFP filter view the film
using 10x or 40 x. The 100x objective will not focus on the film due to
the requirement of oil immersion.
12. Record fluorescent image using a 30 second exposure time.
167
Procedure A-6.
Antigen Rabbit Anti-IgG Crosslinking of 3-aminopropyltrimethoxysilane
(APS), Bis[sulfosuccinimidyl]suberate (BS3), Rabbit IgG Functionalized
Diatom Frustules
Debra Gale (revised May 4, 2008)
Specialized Equipment
 Eppendorf 5414 Microcentrifuge
 2.0 mL microcentrifuge tube
 Barnstead Thermolyne Lab Quake Rotisserie, 8.0 rpm, Model # 415110
 Leica DMIL microscope
 GFP filter for Leica DMIL microscope (excitation: 470/40 nm,
emission: 525/50 nm)
 35 mm fluorophore dish (World Precision Instruments, # FD35-100)
 Sterile 20-200 μL pipette tips
 Sterile 100-1000 μL pipette tips
Reagents
 Sterile Phosphate buffered saline (PBS), pH = 7.5 (4.3633 g/L
monosodium phosphate monohydrate, 18.3263 g/L disodium phosphate
heptahydrate)
 ImmunoPure Goat Anti-Rabbit IgG, (H+L), MW = 150,000 g/mol, 2.4
mg/mL (Pierce # 31210)
 Goat Anti-Rabbit IgG (H + L), DyLight 488 Conjugated (1mg/ml)
(Pierce # 35552), excitation/emission: 493 nm/ 518 nm
 Anti-Rabbit IgG (whole molecule)–Gold antibody produced in goat
affinity isolated antibody, aqueous glycerol suspension, 10 nm
(colloidal gold) (Sigma # G7402-.4ML)
 ImmunoPure Goat Anti-Human IgG (H+L) (Pierce # 31130)
 Anti-Human IgG (whole molecule)−FITC antibody produced in goat,
10.7 mg/mL (Sigma # F3512-1ML)
Procedure
Antigen: ImmunoPure Goat Anti-Rabbit IgG, (H+L), MW = 150,000 g/mol,
2.4 mg/mL (Pierce # 31210)
168
1. Add 0.68 mL of ImmunoPure Goat Anti-Rabbit IgG to 1 mL (10
mg/mL) APS, BS3, Rabbit IgG functionalized diatom frustules in PBS
in a microcentrifuge tube using a pipette fitted with a sterile tip.
2. Rotate for 2 hours on the Lab Quake Rotisserie.
3. Centrifuge at 1,000 rpm (82 g) for 10 minutes.
4. Pipette off supernatant.
5. Replace supernatant with fresh PBS solution.
6. Repeat steps 3-5, two times to remove excel antigen in solution.
Procedure
Antigen: Goat Anti-Rabbit IgG (H + L), DyLight 488 Conjugated (1mg/ml)
(Pierce # 35552), excitation/emission: 493 nm/ 518 nm
1. Add 0.04 mL of Goat Anti-Rabbit IgG (H + L), DyLight 488
Conjugated to 1.0 mL (0.25 mg/mL) APS, BS3, Rabbit IgG
functionalized diatom frustules in PBS in a microcentrifuge tube using
a pipette fitted with a sterile tip.
2. Rotate for 2 hours on the Lab Quake Rotisserie.
3. Centrifuge at 1,000 rpm (82 g) for 10 minutes.
4. Pipette off supernatant.
5. Replace supernatant with fresh PBS solution.
6. Repeat steps 3-5, two times to remove excess antigen in solution.
7. Pipette 100 µL of Dylight 488 labeled biofunctionalized frustules into a
35 mm fluorophore dish.
8. View at 100x, oil immersion using the Leica DMIL microscope
equipped with the GFP filter.
9. Record fluorescent image using a 30 second collection time.
Procedure
Antigen: Anti-Rabbit IgG (whole molecule)–Gold antibody produced in goat
affinity isolated antibody, aqueous glycerol suspension, 10 nm (colloidal
gold) (Sigma # G7402-.4ML)
1. Add 0.084 mL of Anti-Rabbit IgG (whole molecule) 10 nm gold
particle labeled to 1.0 mL (0.01 mg/mL) APS, BS3, Rabbit IgG
functionalized diatom frustules in PBS in a microcentrifuge tube using
a pipette fitted with a sterile tip.
2. Rotate for 2 hours on the Lab Quake Rotisserie.
3. Centrifuge at 1,000 rpm (82 g) for 10 minutes.
4. Pipette off supernatant.
169
5. Replace supernatant with fresh PBS solution.
6. Repeat steps 3-5, two times to remove excel gold labeled antigen in
solution.
7. Pipette 50 µL of gold labeled frustules onto a lacey carbon TEM grid
for imaging.
Procedure
Antigen: ImmunoPure Goat Anti-Human IgG (H+L), 2.4 mg/mL (Pierce #
31130)
1. Add 0.68 mL of ImmunoPure Goat Anti-Human IgG (H+L) (Pierce #
31130) to 1.0 mL (10 mg/mL) APS, BS3, Rabbit IgG functionalized
diatom frustules in PBS in a microcentrifuge tube using a using a
pipette fitted with a sterile tip.
2. Rotate for 2 hours on the Lab Quake Rotisserie.
3. Centrifuge at 1,000 rpm (82 g) for 10 minutes.
4. Pipette off supernatant.
5. Replace supernatant with fresh PBS solution.
6. Repeat steps 3-5, two times to remove excess antigen in solution.
Procedure
Antigen: Anti-Human IgG (whole molecule)−FITC antibody produced in
goat, 10.7 mg/mL (Sigma # F3512-1ML), excitation: 495 nm, emission: 519
nm
1. Add 0.004 mL of Anti-Human IgG (whole molecule)-FITC to 1.0 mL
(0.25 mg/mL) APS, BS3, Rabbit IgG functionalized diatom frustules in
PBS in a microcentrifuge tube using a pipette fitted with a sterile tip.
2. Rotate for 2 hours on the Lab Quake Rotisserie.
3. Centrifuge at 1,000 rpm (82 g) for 10 minutes.
4. Pipette off supernatant.
5. Replace supernatant with fresh PBS solution.
6. Repeat steps 2-5, two times to remove excess antigen in solution.
7. Pipette 100 µL of FITC labeled biofunctionalized frustules in a 35 mm
fluorophore dish.
8. View at 100x, oil immersion using the Leica DMIL microscope
equipped with the GFP filter.
9. Record fluorescent image using a 30 second collection time.
170
Procedure A-7.
Harrison’s and Guillard’s f/2 enrichment Artificial Seawater Medium
Debra Gale (revision 02/08/09)
Reference:
Harrison, P.J., Waters, R.E. Taylor, F.J.R. (1980) A Broad Spectrum Artificial
Seawater Medium for Coastal and Open Ocean Phytoplankton, Journal of
Phycology. 16, 28-35.
Date Prepared: __________________
Prepared By: ____________________
Stock Solution 1: Salt Solution I, 2L, 1x
1. Measure out the following masses of components and add to 2 L
volumetric flask. Record the lot number, expiration date, and actual mass
added.
 287.9268 g Sodium Chloride (NaCl) FW 58.44 – lot
#:____________________ Mass#:____________________
 48.2371 g Sodium Sulfate (Na2SO4) FW 142.04 – lot
#:____________________ Mass#:____________________
 8.1392 g Potassium Chloride (KCl) FW 74.55 – lot
#:____________________ Mass#:____________________
 2.3642 g Sodium bicarbonate (NaHCO3) FW 84.01 – lot
#:____________________ Mass#:____________________
 1.1726 g Potassium Bromide (KBr) FW 119.00 – lot
#:____________________ Mass#:____________________
 0.3126 g Boric Acid (H3BO3) FW 61.83 – lot
#:____________________ Mass#:____________________
 0.0380 g Sodium Fluoride (NaF) FW 41.99 – lot
#:____________________ Mass#:____________________
2. Fill with DI to the 2 L mark
3. Mix with a stir bar until all salts have dissolved.
4. Pour solution into a 2 L imax bottle and label “Stoc Solution 1, 1x”.
171
Stock Solution 2: Salt Solution II, 2L, 5x
1. Measure out the following masses of components and add to 1 L
volumetric flask. Record the lot number, expiration date, and actual mass
added.
 364.937 g Magnesium Chloride Tetrahydrate (MgCl2 6H2O) FW
203.30 – lot #:____________________
Mass#:____________________
 51.1347 g Calcium Chloride Di hydrate (CaCl2 2H2O), FW 147.01 – lot
#:____________________ Mass#:____________________
 0.829 g Strontium Chloride Hexahydrate (SrCl2 6H2O) FW 266.62– lot
#:____________________ Mass#:____________________
2. Fill with DI to the 2 L mark
3. Mix with a stir bar until all salts have dissolved.
4. Pour solution into a 2 L imax bottle and label “Stoc Solution 2, 5x”.
Stock Solution 3: Harrison’s (Major Nutrient I + Major Nutrient II) +
Guillard (f/2 medium), 2L, 10x
1. Measure out the following masses of components and add to 2 L
volumetric flask. Record the lot number, expiration date, and actual
mass added.


90.426 g Sodium Nitrate (NaNO3) FW 84.99 – lot
#:____________________ Mass#:____________________
6.668 g Sodium Phosphate Monobasic (NaH2PO4 H2O) FW
137.99– lot #:____________________
Mass#:____________________
2. Fill with DI to the 2 L mark
3. Mix with a stir bar until all salts have dissolved
5. Pour solution into a 2 L imax bottle and label “Stoc Solution 3, 10x”.
172
Stock Solution 4: Silica (Sodium Metasilicate), 200 mM Si
1. Measure out the following masses of components and add to 1 L
volumetric flask. Record the lot number, expiration date, and actual
mass added.

2.
3.
4.
5.
42.428 g Sodium Metasilicate (Na2SiO3 5H2O) FW 212.14– lot
#:____________________ Mass#:____________________
Fill with DI to the 1 L mark
Mix with a stir bar until all salts have dissolved.
Pour “Stoc Solution 4” into a polypropylene bottle.
Pour solution into a 1 L polypropylene bottle and label “Stoc Solution
4, 200 mM Si”.
Stock Solution 5: Harrison’s (Metal Stock I + Metal Stock II) + f/2 trace
metals
Zinc Sulfate Heptahydrate Superstock
1. Measure out the following masses of components and add to 1 L
volumetric flask. Record the lot number, expiration date, and actual
mass added.

0.01 g Zinc Sulfate Heptahydrate (ZnSO4 7H2O) FW 287.56– lot
#:____________________ Mass#:____________________
2. Fill with DI to the 1 L mark
3. Mix with a stir bar until all salts have dissolved.
4. Pour solution into a 1 L imax bottle and label “Zinc Sulfate
Heptahydrate, 0.01 g/L”.
Sodium Molybdate Superstock
173
1. Measure out the following masses of components and add to 1 L
volumetric flask. Record the lot number, expiration date, and actual
mass added.

0.01 g Sodium Molybdate Dihydrate (Na2MoO4 2H2O) FW
241.95– lot #:____________________
Mass#:____________________
2. Fill with DI to the 1 L mark
3. Mix with a stir bar until all salts have dissolved.
4. Pour solution into a 1 L imax bottle and label “Sodium Molybdate
Dihydrate, 0.01 g/L”.
Copper Sulfate Pentahydrate Superstock
1. Measure out the following masses of components and add to 1 L
volumetric flask. Record the lot number, expiration date, and actual
mass added.

0.01 g Copper Sulfate Pentahydrate (CuSO4 5 H2O) FW 249.69– lot
#:____________________ Mass#:____________________
2. Fill with DI to the 1 L mark
3. Mix with a stir bar until all salts have dissolved.
4. Pour solution into a 1 L imax bottle and label “Copper Sulfate
Pentahydrate, 0.01 g/L”.
Sodium Selenite Superstock
1. Measure out the following masses of components and add to 1 L
volumetric flask. Record the lot number, expiration date, and actual
mass added.

0.005 g Sodium Selenite (Na2SeO3) FW 172.94 – lot
#:____________________ Mass#:____________________
2. Fill with DI to the 1 L mark
3. Mix with a stir bar until all salts have dissolved.
4. Pour solution into a 1 L imax bottle and label “Sodium Selenite, 0.005
g/L”.
Cobalt (II) Sulfate Heptahydrate Superstock
174
1. Measure out the following masses of components and add to 1 L
volumetric flask. Record the lot number, expiration date, and actual
mass added.

_____g Cobalt (II) Sulfate Heptahydrate (CoSo4 H2O) – lot
#:____________________ Mass#:____________________

Fill with DI to the 1 L mark
Mix with a stir bar until all salts have dissolved.
Pour solution into a 1 L imax bottle and label “Cobalt (II) Sulfate
Heptahydrate, _____ g/L”.


Nickel Chloride Hexahydrate Superstock
2. Measure out the following masses of components and add to 1 L
volumetric flask. Record the lot number, expiration date, and actual
mass added.

0.01 g Nickel Chloride Hexahydrate (NiCl2 6H2O) FW 237.69 – lot
#:____________________ Mass#:____________________
3. Fill with DI to the 1 L mark
4. Mix with a stir bar until all salts have dissolved.
5. Pour solution into a 1 L imax bottle and label “Nic el Chloride
Hexahydrate, 0.01 g/L”.
Stock Solution 5 Preparation, 2L, 10x:
1. Measure out the following components in a 1 L volumetric flask.
Record the lot number, expiration date, and actual mass added.





0.212 g Iron(III) Chloride Hexahydrate (FeCl3 6H2O) – lot
#:____________________ Mass#:____________________
0.308 g Ethylenedinitrol tetraacetic Acid Disodium Salt
(C10H14O8N2Na2 2H2O) – lot #:____________________
Mass#:____________________
0.0204 g Manganese (II) Sulphate Tetrahydrate (MnSO4 4H2O) – lot
#:____________________ Mass#:____________________
4.86 mL of “Zinc Sulfate Heptahydrate Superstoc ”
138.8 mL of “Cobalt (II) Sulfate Heptahydrate Superstoc ”
175




17.36 mL of “Sodium Molybdate Superstoc ”
22.62 mL of “Copper Sulfate Pentahydrate”
0.658 mL of “Sodium Selenite”
2.84 mL of “Nic el Chloride Hexahydrate Superstoc ”
2. Fill to the 2 L mark on a volumetric flask.
3. Mix with a stir bar until all salts have dissolved
3. Pour solution into a 2 L imax bottle and label “Stoc Solution 5, 20x”.
Stock Solution 6: Vitamin Stock (Harrison + f/2)
Thiamine HCl Superstock
1. Measure out the following components in a 1 L volumetric flask.
Record the lot number, expiration date, and actual mass added.

0.01 g Thiamine HCl (C12H18N4OSCl) – lot #:____________________
Mass#:____________________
2. Fill to the 1 L mar and label as “Thiamine HCl Superstoc ”
Biotin Superstock
1. Measure out the following components in a 1 L volumetric flask.
Record the lot number, expiration date, and actual mass added.

0.005 g Biotin (C10H16N2O3S) – lot #:____________________
Mass#:____________________
2. Fill to the 1 L mar and label as “Biotin Superstoc ”
Vitamin B12 Superstock
1. Measure out the following components in a 1 L volumetric flask.
Record the lot number, expiration date, and actual mass added.

0.010 g Vitamin B12 (C63H88N14CoO14P) – lot
#:____________________ Mass#:____________________
176
2. Fill to the 1 L mar and label as “Vitamin B12 Superstoc ”.
Stock Solution 6 Preparation:
1. Measure out the following components in a 1 L volumetric flask.
Record the lot number, expiration date, and actual mass added.



8.616 mL “Thiamine HCl Superstoc ”
0.005 mL “Biotin Superstoc ”
0.1285 mL “Vitamin B12 Superstoc
2. Fill to the 1 L mark.
3. Sterile filter the solution according to the “Sterile Filter Procedure”.
4. Store in 15 mL aliquots in the freezer in sterile centrifuge tubes.
Harrison’s Artificial Seawater Medium + Guillard’s f/2 enrichment
1. Add the following components to make 1 L artificial seawater medium.
o
o
o
o
o
o
140 mL Stock Solution 1-1x
50 mL Stock Solution 2-5x, 200 mL DI
90 mL Stock Solution 3-10x, 90 mL DI
5.5 mL Stock Solution 4
50 mL Stock Solution 5-10x, 450 mL DI
4.55 mL Stock Solution 6
177
Procedure A-8.
Avidin-Fluorescein (FITC) Functionalization of Biotinylated Diatom
Frustules
Debra Gale (revised December 31, 2006)
Specialized Equipment
 Eppendorf 5414 microcentrifuge
 1.5 mL black microcentrifuge tube
 New Brunswick Scientific Co. Rotator, model # TC-8
 Leica DMIL microscope
 DAPI filter (ex/em: 340-380 nm/425 longpass)
 35 mm fluorophore dish (World Precision Instruments, # FD35-100)
Reagents
 Immunopure Avidin, Fluorescein (FITC) Conjugated (Pierce
Biotechnology, #21221)
 Phosphate buffered saline (PBS) prepared according to Rorrer Lab
Procedure
Procedure
1. Add 0.08 mL of 5 mg/mL FITC-avidin solution to 0.25 mL (1 mg/mL)
biotin functionalized phosphate buffered saline suspended diatom frustules
in a black microcentrifuge tube to protect the FITC from light.
2. Incubate the reaction on a Lab Quake Rotisserie at 8 rotations per minute
for 1 hour at room temperature.
3. Microcentrifuge the suspension for 8 minutes at 1000 rpm.
4. Pipette off the supernatant and discard.
5. Replace the supernatant with 1 mL of phosphate buffered saline.
6. Repeat steps 3 through 5 ten times to remove un-reacted avidin-FITC.
7. Transfer 20 μL of the avidin-FITC functionalized biotinylated diatom
frustules to a 35 mm fluorodish
8. Use the Leica microscope equipped with a DAPI filter to validate the
presence of the fluoroprobe FITC which indicates that the biotin-avidin
was functionalized on the diatom frustule surface.
178
Procedure A-9.
Biotin Functionalization of Amine Functionalized Diatom Frustules
Debra Gale (revised December 31, 2006)
Specialized Equipment
 Eppendorf 5414 microcentrifuge tube
 1.5 mL black microcentrifuge tube
 New Brunswick Scientific Co. Rotator, model # TC-8, 8 rpm
 Barnstead Thermolyne Lab Quake Rotisserie, 8.0 rpm, Model # 415110
Reagents
 EZ-Link Sulfo-NHS-Biotin
(sulfosuccinimidobiotin) (Pierce Biotechnology, #21217)
 Reagent grade ethanol
 Phosphate buffered saline (PBS) pH 7.2
Procedure
1. Suspend 1 mg of amine functionalized diatom frustules in phosphate
buffered saline.
2. Wash diatom suspension with phosphate buffered saline by
microcentrifuging for 8 minutes at 1000 rpm.
3. Pipette off the supernatant and discard.
4. Replace the supernatant with 1 mL of phosphate buffered saline.
5. Repeat steps 2-4 four times.
6. Dissolve 2.2 mg EZlink sulfo NHS Biotin in a 0.5 mL ultrapure water to
prepare a 10 mM Biotin solution.
7. Add 0.25 mL (10 mM) Biotin solution and 0.25 mL of 1 mg/mL diatom
frustules in phosphate buffered saline into a microcentrifuge tube.
9. Incubate the reaction on a Lab Quake Rotisserie at 8 rotations per minute
for 1 hour at room temperature.
8. Repeat steps 2 through 4 three times to remove un-reacted biotin.
179
Procedure A-10.
Bis[sulfosuccinimidyl]suberate (BS3) Crosslinking of 3aminopropytrimethoxysilane (APS) Functionalized Diatom Frustules
Debra Gale (revised May 4 ,2008)
Specialized Equipment
 Eppendorf 5414C Microcentrifuge
 2.0 mL microcentrifuge tube
 Barnstead Thermolyne Lab Quake Rotisserie, 8.0 rpm, Model # 415110
Reagents
 Sterile Phosphate buffered saline (PBS), pH = 7.5 (4.3633 g/L
monosodium phosphate monohydrate, 18.3263 g/L disodium phosphate
heptahydrate)
 Bis[sulfosuccinimidyl]suberate (BS3), MW=572.42 g/mol,
C16H18N2Na2O14S2 (Pierce Biotechnology # 21580)
 10 mg 3-aminopropytrimethoxysilane (APS) functionalized diatom
frustules prepared from the Rorrer lab procedure
Procedure
1. Wash 10 mg/mL of 1.5 mL APS functionalized frustules in PBS buffer
by centrifuging at 1,000 rpm (82 g) for 10 minutes in a microcentrifuge
tube.
2. Remove the supernatant.
3. Replace supernatant with 1.0 mL PBS buffer.
4. Repeat steps 1-3, three times for a total of four washes to remove any
excess APS or ethanol from the previous reaction step.
5. Add 3.30 mg BS3 to APS functionalized frustules in the PBS solution
in the microcentrifuge tube.
6. Invert five times to mix and dissolve the BS3. Do not shake violently.
7. Rotate on the Lab Quake Rotisserie for 15 minutes.
8. Centrifuge BS3 functionalized frustule solution for 5 minutes at 1,000
rpm (82 g).
9. Remove supernatant.
10. Quickly proceed with additional functionalization steps as BS3
hydrolyzes quickly.
180
** Representatives at Pierce Biotechnology suggest that after the BS3 has
reacted for 15 minutes, the antibody can be directed added to the reaction
mixture without washing away BS3 or it can be washed away.
181
Procedure A-11.
Bis[sulfosuccinimidyl]suberate (BS3) Crosslinking of 3aminopropytrimethoxysilane (APS) Functionalized Diatom Frustule Films
on Round Coverslips
Debra Gale (revised August 25, 2009)
Specialized Equipment
 1.5 mL Microcentrifuge Tube
 6 Well VWR Polystyrene Plate
 VWR Orbital Shaker, 50 rpm
Reagents
 Phosphate buffered saline made the Rorrer lab procedure Phosphate
Buffered Saline Preparation, pH 7.6, 0.01M Phosphate, 0.15 M NaCl
08-25-09
 Bis[sulfosuccinimidyl]suberate (BS3), MW=572.42 g/mol,
C16H18N2Na2O14S2 (Pierce Biotechnology # 21580)
Sample
 1- APS functionalized diatom frustule film on a round coverslip, 0.060
mg, approximately 5 mm in diameter
Procedure
1. Add 1.8 mL of PBS to one well of the six well plate.
2. Gently place the APS functionalized diatom frustule thin film (biosilica
side down) on the surface of the PBS filled six well plate. The surface
tension of the PBS solution will prevent the coverslip from sinking
such that the entire APS-BS3 reaction takes place with the coverslip
floating on the surface.
3. Add 0.200 mL of a 0.0657 mg/mL solution of BS3 in PBS in a
microcentrifuge tube to the PBS filled well which has the coverslip
floating on the PBS surface.
4. Allow to rotate on the shaker at 50 rpm for 15 minutes.
5. Remove coverslip and transfer to a second well containing fresh PBS
solution to wash, or if continuing to another reaction step then place
directly in second well with antibody solution.
6. Quickly proceed with additional functionalization steps as BS3
hydrolyzes quickly.
182
Procedure A-12.
Coscinodiscus wailesii -Rh123 Sample Preparation for Deposition on PAH
Polymer
Debra Gale (revised October 28, 2008)
Specialized Equipment
 Eppendorf 5414 Microcentrifuge
 2.0 mL microcentrifuge tube
 Barnstead Thermolyne Lab Quake Rotisserie, 8.0 rpm, Model # 415110
 Leica DMIL microscope
 GFP filter for Leica DMIL microscope (excitation: 470/40 nm,
emission: 525/50 nm)
 35 mm fluorophore dish (World Precision Instruments, # FD35-100)
 Sterile 20-200 μL pipette tips
 Sterile 500-5000 μL pipette tips
 500-5000 μL Pipette
 20-200 μL Pipette
 4, 250 mL Erlenmeyer flasks with foam stoppers
 50 mL red line graduated cylinder
 100 μm filter mesh
Reagents
 Cold Methanol (CAS # 67-56-1, MW 32.04 g/mol) which has been
stored at -4ºC
 Deionized water
 5 wt% Sodium Dodecyl Sulfate (CAS# 151-21-3, MW 288.38 g/mol)
and 100 mM EDTA (ethylenediaminetetraacetic, CAS # 60-00-4, MW
292.24 g/mol) in Deionized water
 Harrison’s Artificial Seawater Medium enriched with f/2 nutrients and
0.65 mM silicon as Na2SiO3 (Harrison, P.J. et al., 1980)
 3.0 mg of Rhodamine 123 in 50 mL deionized water (60 mg Rh123/L),
(Sigma # R8004-25 mg, MW 380.82 g/mol, CAS # 62669-70-9,
excitation 505 nm/ emission 534 nm)
 Coscinodiscus wailesii, CCMP Bigelow Laboratory for Ocean Sciences
#2513
183
Procedure
1. Measure with the 50 mL red line graduated cylinder 40 mL Harrison’s
Artificial Seawater Medium into each Erlenmeyer flask and cover with
foam stopper.
2. Autoclave Erlenmeyer flasks with artificial seawater for 25 minutes.
3. Remove flasks from autoclave and allow to cool and equilibrate with
the atmosphere for 24 hours.
4. Add 1.67 mL of the 60 mg/L Rhodamine 123 solution to each of the
flasks containing artificial seawater medium. This will yield a final
Rh-123 concentration in the flasks of 2 mg/L, also used by Brzezinski,
M.A. and Conley, D.J. (1994).
5. Pipette 10 mL of a 2,925 cells/mL Coscinodiscus wailessi culture into
each Rhodamine containing flask to yield a calculated initial cell
density of 585 cells/mL. The actual measured cell density was
433±175 cells/mL.
6. Place flasks in incubator at 22ºC with a L/D cycle of 14/10 hrs.
7. Swirl Coscinodiscus flasks for 5 seconds each day for seven days.
8. The final cell density measured was 2,100±500 cells/mL.
9. Pipette with the 5000 μL pippetter the contents of one CoscinodiscusRh123 culture flas through the 100 μm filter mesh to remove frustules
fragments and residual free floating cellular material.
10. Rinse the whole frustules from the filter mesh using 10 mL of cold
methanol into a 50 mL beaker.
11. Pipette whole frustules from the methanol suspension into 5, 2 mL
microcentrifuge tubes.
12. Centrifuge the frustules for 5 minutes at 900 rpm.
13. Remove the tube from the centrifuge and pipette off the supernatant.
14. Combine all of the frustules into one microcentrifuge tube and fill with
cold methanol.
15. Centrifuge the frustules for 5 minutes at 900 rpm.
16. Remove the methanol supernatant and replace with deionized water.
17. Centrifuge the frustules for 5 minutes at 900 rpm.
18. Remove the supernatant and replace with 5 wt% SDS and 100 mM
EDTA solution.
19. Place SDS and frustule suspension in the microcentrifuge tube on the
rotator and allow to rotate at 8 rotations/minute for 24 hours. The SDS
and EDTA will remove cellular organic material, but will not quench
the fluorescence of the Rh123.*
20. Centrifuge the frustules for 5 minutes at 900 rpm.
21. Remove the supernatant and replace with deionized water.
184
22. Repeat steps 19 and 20 four times to remove SDS and EDTA mixture
and cellular organics in solution.
23. Repeat step 19.
24. Remove the supernatant and replace with 0.5 M NaCl solution. The
diatoms must be in a NaCl solution for deposition and attachment to the
positively charged polyelectrolyte surface.
25. Centrifuge the frustules for 5 minutes at 900 rpm.
26. Remove the supernatant and replace with 0.5 M NaCl solution.
27. To validate Rh123 incorporation remove 50 μL from the frustule
suspension in 0.5 NaCl solution and place in a 35 mm fluorophore dish.
28. View the Coscinodiscus frustules using the Leica DMIL microscope
equipped with a GFP filter with an integration time of 10 seconds.
* The hydrogen peroxide treatment method to remove organic cellular material
will quench the fluorescence of the Rh123.
References
Brzezinski, M.A. and Conley, D.J. (1994) Silicon Deposition During the Cell
Cycle of Thalassiosira Weissflogii (Bacillariophyceae) Determined Using Dual
Rhodamine 123 and Propidium Iodide Staining, Journal of Phycology. 30, 4555.
Harrison, P.J., Waters, R.E. Taylor, F.J.R. (1980) A Broad Spectrum Artificial
Seawater Medium for Coastal and Open Ocean Phytoplankton, Journal of
Phycology. 16, 28-35.
185
Procedure A-13.
Deposition of Polyectrolyte Multilayer on Microscope Slide
Process developed by Wei Wang, procedure write-up by Clayton Jeffryes
(revised Jan 28, 2008)
Revised by Debra Gale, February 28, 2009
Unique Materials

Microscope slide cut into 1 in x 0.5 in pieces (borosilicate glass)

100 ml beakers (Dipping Containers)
Specialized Equipment

Nitrogen cylinder with gas nozzle
Reagents

Deionized water

Acetone

Methanol

0.5 M NaCl (Filtered, pore size = 0.45 micron)

Poly(styrene sulfonic acid) sodium salt (PSS), MW=70,000, CAS# 2570418-1

Poly(allylamine hydrochloride) (PAH), MW=15,000, CAS#71550-12-4

0.5 M NaOH in 30% H2O2
Sample

Coscinodiscus wailesii frustules which have been washed in 5% SDS (See
Procedure “SDS Removal of Organics”) and dispersed in 0.5 M NaCl
Procedure
1. Prepare 40 ml of 3 mg ml-1 PSS in 0.5 M NaCl solution in deionized water.
2. Prepare 40 ml of 1.35 mg ml-1 PAH in 0.5 M NaCl solution in deionized
water.
3. Submerge microscope slide in NaOH/H2O2 solution in a 100 ml beaker and
sonicate for 10 min.
4. Remove microscope slide and rinse with DIH2O, methanol and acetone.
5. Repeat steps 3 and 4 two more times for a total of three cleanings.
186
6. Dry microscope slide by spraying with pressurized nitrogen from the
nitrogen cylinder.
7. Put 40 ml of PSS and PAH solution into separate 100 ml beakers.
8. Dip glass into PAH solution for 20 minutes.
9. Rinse for 30 seconds with DIH2O and blow dry with compressed nitrogen
10. Dip glass into PSS solution for 20 minutes.
11. Repeat step 9.
12. Repeat steps 8-11 three times for a total of 4 bilayers.
13. Print PAH in desired pattern on the PSS top layer of the PSS/PAH
multilayers.
14. Rinse for 30 seconds with DIH2O and blow dry with compressed nitrogen.
15. Deposit 1 mL of the Coscinodiscus frustule solution in NaCL onto the
glass slide and allow to adsorb for 20 minutes. Target the frustule
concentration to form a monolayer on the printed PAH.
16. Rinse for 30 seconds with DIH2O and blow dry with compressed nitrogen.
187
Procedure A-14.
APS Functionalized Diatom Frustule Film Preparation for PL Sensitivity
Study
Debra Gale (revised June 19, 2008)
Specialized Equipment
 Eppendorf 5414C microcentrifuge
 VWR* Micro Cover Glasses, Round, No. 1, 18 mm diameter, (23/32) inch
diameter (#48380-046) (micro coverslips 23 mm in diameter do not fit on
the PL sample holders and are currently the only coverslips sold in Chemstores in Gilbert hall)
 200 μL – 1000 μL Pipette and tips
 One piece green engineering paper which has 5 mm boxes
 Translucent plastic surface through which the engineering paper can be
viewed underneath (clear plastic clipboard works well)
Sample
 1 mg/mL APS functionalized H2O2 treated diatom cells (frustules)
suspension in deionized water in a 1.5 mL microcentrifuge tube
 This sample can also be 3-aminopropyltrimethoxysilane (APS)
functionalized diatom biosilica
Procedure
1. Arrange one round coverslip on the translucent surface which is on the grid
side of the engineering paper with the grid in the center of the coverslip.
2. Invert frustule suspension in the microcentrifuge to ensure a homogenous
solution.
3. Pipette 20 μL of the frustule suspension onto the center of the glass
coverslip.
4. Allow droplet to dry as a white film.
5. Again, pipette 20 μL of the frustule suspension onto the center of the glass
coverslip.
6. Allow droplet to dry as a white film.
7. Again, pipette 20 μL of the frustule suspension onto the center of the glass
coverslip for a total of 60 μL solution which yields 60 μg of biosilica on
the coverslip as a 5 mm in diameter frustule film.
188
8. Allow droplet to dry as a white film.
9. If the sample used is APS functionalized diatom biosilica, then heat the
dried APS-biosilica thin film for 1 hr at 90°C in a pre-heated furnace.
189
Procedure A-15.
Fluorescamine Conjugation of Amine Functionalized Diatom Frustule
Films on Round Coverslips
Debra Gale (revised June 22, 2008)
Specialized Equipment
 Leica DMIL microscope
 DAPI filter (ex/em: 340-380 nm/425 longpass)
 6 Well Polystyrene Plate
 VWR shaker, 60 rpm
Reagents
 Fluorescamine FluoroPureTM grade (Molecular Probes, # F-20261, MW
278.26 g/mol)
 Reagent grade anhydrous methanol
Sample
60 µg of APS functionalized diatom biosilica deposited as a 5 mm in diameter
film on a round glass coverslip prepared according to the Rorrer lab procedure
Procedure
1. Add 1.5 mL of ethanol to one well of a six well plate.
2. Carefully place coverslip ( on the liquid interface, film side down. The
surface tension of the ethanol will prevent the coverslip from sinking.
3. Dissolve 0.5 mg fluorescamine in 1 mL anhydrous methanol in a black
microcentrifuge tube, protecting this reagent from light.
4. Vortex the solution for 30 seconds to ensure the fluorescamine has
dissolved in the methanol.
5. Add 0.06 mL of 0.5 mg/mL fluorescamine to the ethanol filled well.
6. Cover with foil to prevent photobleaching of the fluorescamine.
7. Set the shaker at 60 rpm to ensure homogeneous mixing.
8. Allow the fluorescamine-amine reaction to take place for 10 minutes.
9. The following steps must be done in under low light.
10. Remove the coverslip with tweezers.
190
11. Dip the coverslip in deionized (DI) water to wash away excess unreacted
fluroescamine.
12. Again,Dip the coverslip in DI water to wash away excess unreacted
fluorescamine.
13. Place the coverslip film side up on a piece of paper and allow to dry. The
next step can be completed prior to drying.
14. Use a Leica DMIL microscope equipped with a DAPI filter and 30 second
exposure time to validate the presence of fluorescamine on the frustule.
Fluorescamine that has not reacted with a primary amine will not fluoresce
because it will hydrolyze.
191
Procedure A-16.
Fluorescamine Conjugation of Amine Functionalized Diatom Frustules
Debra Gale (revised May 4, 2008)
Specialized Equipment
 Leica DMIL microscope with fluorescent light source
 DAPI filter (ex/em: 340-380 nm/425 longpass)
 1.5 mL black microcentrifuge tube
 VWR Vortex Genie 2
 35 mm fluorophore dish (World Precision Instruments, # FD35-100)
Reagents
 Fluorescamine FluoroPureTM grade (Molecular Probes, # F-20261, MW
278.26 g/mol)
 Reagent grade anhydrous methanol, CH3OH, 32.04 g/mol
Sample
0.01 mg/mL suspension of 3-aminopropyltrimethoxysilane (APS)
functionalized diatom frustules in ethanol
Procedure
1. Dissolve 0.50 mg fluorescamine in 1 mL anhydrous methanol in a black
microcentrifuge tube, protecting this reagent from light.
2. Vortex the solution for 30 seconds to ensure the fluorescamine has
dissolved in the methanol.
3. In a separate black microcentrifuge tube, combine 0.010 mL (0.5 mg/mL)
Fluorescamine and 1.0 mL (0.01 mg/mL) diatom suspension in ethanol.
4. Vortex this solution for two minutes to allow fluorescamine to react with
primary amines located on the APS functionalized diatom frustule.
5. Protecting from light, transfer 20 μL of the fluorescamine labeled diatom
solution to a 35 mm fluorophore dish.
6. Use a Leica DMIL microscope equipped with a DAPI filter and 30 second
exposure time to validate the presence of fluorescamine on the frustule.
Fluorescamine that has not reacted with a primary amine will not fluoresce.
The best fluorescent images are acquired when the frustules are in solution.
However, the ethanol will evaporate quickly under the microscope light so
imaging must be done quickly.
192
Procedure A-17.
Micro Raman Spectroscopy Analysis of Diatom Biosilica
Debra Gale (revised May June 3, 2011)
Unique Materials

Calcium fluoride window (Edmund Optics, 5mm Dia. Calcium Fluoride
Window, Uncoated , #NT48-853)
Specialized Equipment

Horiba Jobin Yvon LabRam HR Raman microscope
Sample

Diatom frustules in methanol (1 mg frustules/ mL MeOH)
Sample Preparation

Deposit 10 µL of frustule solution on CaF window and allow to dry. The
frustule concentration in MeOH can be adjusted to yield dispersed single
frustules on the CaF substrate. Due to the CaF properties, the pipette tip
can be used to spread the frustules out on the substrate while the methanol
in evaporating.
Procedure
Raman Measurement
1. Turn on the 532 nm laser. This laser must warm up for 30 minutes prior to
the Raman measurement.
2. Check CCD temperature. It should read -70°C.
3. Open the door on the top of the Raman instrument to visualize Raman
optics.
4. Switch out the 532 nm notch filter and pin.
5. Switch out the 532 nm entrance filter.
6. Replace black optics cover.
7. Close optics door but leave slightly ajar until ready to take a Raman
measurement.
8. Take single point measurement of calibration standard Si and record in log
book.
9. Place CaF frustule sample on glass coverslide and place on microscope
stage.
193
10. Focus on a single frustule with the white bright light and observe through
microscope video.
11. Once a single frustule has been identified, close the Raman optics door
completely to allow 532 nm excitation to enter through the objective and
focus on the frustule.
12. Using the fine focus on the microscope, focus in and out to find the tightest
532 nm laser spot.
13. Set instrument parameters
Raman Instrument Parameters
Laser: 532 nm
Objective: 100x
Grating: 1200 g/mm
Confocal Hole Diameter: 150 µm
Spectral Range collected: 300-1100 cm-1
Spectrum average: 1 (map), 3 (line scan), 3 (single point measurement)
Integration time: 120 sec
Step size for line scan or mapping: 0.350 µm
Neutral Density filter: 0
Estimated theoretical lateral resolution: ~350 nm
Estimated theoretical analysis depth: ~4 µm
14. Proceed with single point or Raman mapping measurement.
a. If mapping with the DUOScan feature, make sure the DUOScan
mirror is pulled out and that the DUOScan video option is checked
in the software. Single point Raman measurements can be done in
the DUOScan mode to check for signal strength.
b. Keep in mind that the spectral range of 300-1100 cm-1 is 2 spectral
windows. This is important because for one single measurement at
120 second integration time, the actual signal collection time is 240
seconds. Under these conditions, line scans take approximately 4
hours and maps take about 20 hours.
Raman Signal Processing
1. The signal background from the raw Raman signal data is subtracted using
the “baseline subtract” function in the Spectra5 Raman processing
software.
2. Raman peaks are identified by the software and this Raman shift and signal
intensity is reported. No post-processing was done other than the baseline
signal subtraction.
194
Procedure A-18.
Phosphate Buffered Saline Preparation, 0.01 M Phosphate, 0.15 M NaCl,
pH 7.6
Debra Gale (revised July 19, 2010)
Unique Materials
 Nalgene Supor® mach V bottle top filter, 50 mm diameter membrane, 0.2
μm pore size
Specialized Equipment
 Autoclave
 1 L Kimax glass bottle
Reagents
 Monosodium phosphate monohydrate, NaH2PO4·H2O, 137.99 g/mol
 Disodium phosphate heptahydrate, Na2HPO4·7H2O, 268.07 g/mol
 Sodium Chloride, NaCl, 58.44 g/mol
Procedure
1. Combine 2.1453 g monosodium phosphate monohydrate, 22.6337 g
disodium phosphate heptahydrate, and 8.766 g sodium chloride in a 1 L
volumetric flask.
2. Fill the volumetric flask with deionized water up to the mark and dissolve
by stirring on a magnetic stir plate.
3. Measure the pH of the solution. It should be 7.6.
4. Autoclave a 1 L Kimax glass bottle at 121º C for 10 minutes.
5. Filter the phosphate solution into the sterile bottle with the Nalgene Supor®
mach V bottle top filter in the laminar flow hood using asceptic technique.
Resulting pH of the phosphate buffered saline should be 7.6.
Reference
This PBS recipe is used by Pierce Biotechnology for protein
(immunocomplex) cross-linking. There are several recipes in the literature, but
this recipe was used due to its preference by Pierce Biotechnology.
195
Procedure A-19.
Photoluminescence (PL) System Alignment
Debra Gale (revised November 30, 2010)
Unique Materials

Not applicable
Specialized Equipment

337 nm Nitrogen laser, beam steering and sample holder mount, and CCD
spectrometer set to 200 micron slit width

UV safety glasses (wear glasses during the system alignment)
Sample

Biosilica standard (Ni-TQ, 40,000 counts at a 2 second integration time)
Procedure Note: This procedure describes the alignment of the PL system after
the point of assembling the items in the PL system photograph. If the user is
familiar with the system and it starts to drift out of alignment, sometimes a
total re-alignment is not necessary.
Procedure
1. Turn on the 337 nm Nitrogen laser with key.
2. Turn the laser pulse to maximum on the back of the nitrogen laser.
3. Turn on the spectrometer by flipping the back switch on the front of the
instrument.
4. Open the “Spectra Sense” program on the computer. The CCD must be
turned on first otherwise the computer will not recognize the device.
5. Clic on the “spectral” setting.
6. Place the biosilica sample standard in the sample holder.
7. Move beam steering mirrors back to their default position with the 3 point
screws.
8. Adjust height of nitrogen laser to the sample height.
9. Adjust height of all beam steering posts to the sample holder height.\
10. Move the laser (1) so that it lines up the first beam steering mirror (2).
11. Adjust beam steering mirrors (2 and 3) using the 3 point screws to send the
laser beam through the slit (4) such that after going through the slit (4) it
hits the sample mounted on the sample holder. Ensure that the angle and
height are optimally adjusted by visual inspection.
196
12. Adjust the angle of the sample holder (5) to steer the PL emission through
the lens (6). The purpose of the lens is to slightly focus the PL emission
such that it can go through the 200 µm slit on the CCD entrance. The PL
emission from the sample will inherently have a small diameter. To obtain
the smallest PL emission diameter, adjust the PL emission from the sample
to hit the lens slightly off center.
13. If the alignment of the front part of the system was done well, then the
CCD should not have to be moved at all. To check the alignment, collect a
PL signal emission with the CCD from the computer using a 2 second
integration time.
14. Move the filter wheel to an empty slot. This is confirmed visually.
15. Collect a spectrum.
16. The signal of the 337 nm excitation point will show up around 6,000,000
counts if the system is optimally aligned.
17. Adjust the CCD angle and collect a spectrum until the 337 nm excitation
line is reported as 6 million counts.
18. Adjust the filter wheel to the excitation cut-off of 360 nm. This filter
wheel can be verified by collecting a spectrum and visually observing the
360 nm cut off.
19. Collect a spectrum. The PL intensity of the biosilica standard should be
approximately 40,000 counts for a 2 second integration time and a peak
height intensity wavelength of approximately 450 nm.
20. Adjust the alignment of the system if the biosilica standard does not read
approximately 40,000 counts using the 360 nm excitation cut-off filter.
21. Take an image of the biosilica standard emission using the CCD. The most
intense emission spot should be centered in the y direction. If the PL
emission is not centered then adjust the alignment.
Photoluminescence System Equipment labeled by number in the photograph
1. 337 nm Nitrogen Laser:
2. Beam steering mirror
3. Beam steering mirror
4. Beam steering slit
5. Sample holder: 18 mm diameter round glass coverslip fits on delrin coupon
which is held upright by the sample holder (a delrin coupon can also be
loaded with sample within a well)
6. Focusing lens
7. CCD
197
Photograph of
Photoluminescence set up in
Gleeson Laboratory 302
7
2
1
6
4
3
5
Figure A-1. Photograph of the photoluminescence system in Gleeson
laboratory 308.
198
Procedure A-20.
Pinnularia Diatom Deposition on Poly(dimethylsiloxane) (PDMS) Coated
Glass Slide
Debra Gale (revised August 22, 2008)
Specialized Equipment
 Drying Oven
 25 mm x 75 mm glass slide
 20-200 μL pipette with tips
Reagents
 Poly(dimethylsiloxane) (PDMS), 200 fluid, viscosity 5 cSt, 136.2 g/mol,
0.963 g/mL at 25ºC (Sigma # 317667)
 8x104 cells/mL hydrogen peroxide treated Pinnularia frustules in methanol
Procedure
1. Deposit 0.20 mL of PDMS on a glass slide using a pipette.
2. Spread PDMS evenly over the glass slide into a very thin layer using the
edge of a second glass slide or a flat spatula.
3. Place the PDMS covered slide in the drying oven for 30 minutes at 80ºC.
Heating will dehydrate the PDMS into a rubber-like solid.
4. Deposit 0.250 mL of the 8x104 cells/mL Pinnularia frustules in methanol
on the glass slide. The frustule solution in methanol will spread to cover
the whole slide.
5. Again, place the PDMS and Pinnularia frustule covered glass slide in the
oven at 80ºC for 15 minutes to drive off the methanol leaving a disperse
layer of Pinnularia frustules on a PDMS covered glass slide. See the
following image for an example of the coverage.
199
50 μm
Figure A-2. Pinnularia diatom deposition on PDMS coated glass slide.
200
Procedure A-21.
Pinnularia Diatom Film Deposition on Glass Coverslips for Annealing and
Photoluminescence Measurements
Debra Gale (revised September 22, 2008)
Specialized Equipment
 VWR* Micro Cover Glasses, Round, No. 1, 18 mm diameter, (23/32) inch
diameter (#48380-046)
 10 μL – 100 μL Pipette and tips
 One piece green engineering paper
Sample
 1 mg/mL H2O2 treated Pinnularia sp. diatom cells (frustules) suspension in
deionized water in a 1.5 mL microcentrifuge tube
Procedure
1. Arrange one round coverslip on the grid side of engineering paper with one
square in the center of the coverslip.
2. Invert 1 mg/mL frustule suspension in the microcentrifuge to ensure a
homogenous solution.
3. Pipette 20 μL of the frustule suspension onto the center of the glass
coverslip.
4. Allow droplet to dry as a white film.
5. Again, pipette 20 μL of the frustule suspension onto the center of the glass
coverslip.
6. Allow droplet to dry as a white film.
7. Pipette 20 μL of the frustule suspension onto the center of the glass
coverslip, then allow to dry for a total of 60 μL. This preparation will yield
60 μg of Pinnularia biosilica deposited as a film which is 5 mm in
diameter.
8. Once droplet has dried the coverslip with biosilica film can be used as a
sample holder for photoluminescence measurements or XPS
measurements.
201
Procedure A-22.
Preparation of Wetted Cyclotella sp. Biosilica for Photoluminescence
Spectroscopy
Debra Gale (revision January 13, 2008)
Specialized Equipment

Delrin polymer photoluminescence (PL) sample holder with 3 mm x 3 mm
x 1 mm well

VWR Micro Cover Glasses, Round, No. 1, 18 mm in diameter (VWR #
48380-046)

Metal spatula
Sample
1 mg of cyclotella diatom frustules in 1 mL of deionized water within a 1.5 mL
micro-centrifuge tube
Procedure
1. Centrifuge diatom frustule suspension at 1,000 rpm (82 g) for 10 minutes.
2. Remove supernatant so that only pellet of 1 mg biosilica remains in the
centrifuge tube.
3. Scoop out 1.0 mg of frustules from the centrifuge tube with the metal
spatula.
4. Place frustules in the Delron sample holder well.
5. Pat the sample flat with the metal spatula, ensuring that the well is filled.
6. Place a circular cover glass over the sample holder to cover the sample.
202
Procedure A-23.
Preparation of Rabbit Immunoglobulin G (IgG) for Photoluminescence
Spectroscopy
Debra Gale (revision January 19, 2008)
Specialized Equipment

Delrin polymer photoluminescence (PL) sample holder with 3 mm x 3
mm x 1 mm well

VWR Micro Cover Glasses, Round, 18 mm in diameter, No. 1 (VWR
# 48380-046)

Sterile 1 mL needle and syringe
Reagents
 ImmunoPure Rabbit IgG, Whole Molecule (Pierce Biotechnology #
31235),
MW = 150,000 g/mol, 11.7 mg/mL
Procedure
1. Remove 0.004 mL of IgG from the ImmunoPure Rabbit IgG glass bottle
using a sterile 1 mL needle and syringe.
2. Deposit Rabbit IgG into well of delrin sample holder filling slightly above
the top of the well.
3. Remove any air bubbles.
4. Cover the delrin sample holder with a circular glass cover slip.
5. Ensure that there is no air in the well of the sample holder and no void
space.
6. Measure photoluminescence.
7. Check sample holder to ensure that there is still no air under coverslip and
sample well is completely filled.
203
Procedure A-24.
Subculture of Cyclotella Sp.
Debra Gale (revision 01/23/06)
Specialized Equipment

Incubator at 20ºC equipped with lights on a 14/10 L/D photoperiod

16, 500 mL flasks with sponge stoppers

Sterile laminar flow hood

Blue flame for sterilization from Fire-boy

20 mL sterile pipette
Reagents

2 L of Harrison’s Artificial Seawater Medium prepared according to the
Rorrer lab procedure
Procedure
1. Label each of the 500 mL flasks with the following:
Cy - Initials - Cell line - Subculture Number - flask number.
2. Fill each flas with 80 mL of Harrison’s artificial seawater medium.
3. Autoclave 16, 500 mL flasks with sponge stoppers according to the
“Autoclaving_glassware_medium_bioreactors” procedure.
4. Put on gloves.
5. Spray out the inside of the laminar flow hood and gloves with 80% Ethanol
in DI.
6. Wipe out the laminar flow hood with a large sized chem wipe.
7. Prepare flasks for subculture in the laminar flow hood.
8. Remove the sponge stopper of the autoclaved flask and flame the lip of the
flask in the blue flame for approximately 3 seconds.
9. Select two flasks of culture from the incubator to be kept as backups.
10. Place the backup culture flasks in the door of the incubator such that they
are out of the light.
11. Remove four flasks of culture from the incubator and place in the laminar
flow hood.
12. Work the following steps in the laminar flow hood.
13. Flame the lips of the flasks and allow them to cool by convection.
14. Combine the flasks into one and swirl to mix.
15. From the combined culture flasks remove 20 mL of culture medium using
a sterile pipette.
204
16. Transfer the 20 mL of culture medium into one 500 mL flasks which is
filled with 80 mL of sterile seawater medium.
17. Swirl flask to mix.
18. Repeat steps 15 through 17 for all 500 mL flasks.
19. Remove old flasks from the incubator.
20. Add 5 mL of bleach per 100 mL of culture from the last subculture for
disposal.
21. Dispose of old bleached cultures down the drain.
22. Place new subcultured flasks back into the incubator set at 20ºC.
23. Swirl flasks each day for 1-2 seconds.
24. Repeat subculture bimonthly staggering cultures.
205
Procedure A-25.
Thermal Gravimetric Analysis (TGA) of Diatom Biosilica
Debra Gale (revised May 3, 2009)
Unique Materials

Aluminum sample pan
Specialized Equipment

TGA 2950 Thermogravimetric Analyzer, Located in Gleeson 206 cage,
Person in Charge: Skip Rochefort
Sample

Dry hydrogen peroxide treated biosilica which has been under vacuum
conditions for at least 7 days, ~1.5 mg for each TGA run
Procedure
22. Log into the user file located on the desktop of the computer next to the
TGA.
23. Turn on the compressed air at the tank.
24. Check air pressure in tank to ensure that it will not run out during sample
analysis. If air pressure in tank is below 200 psi, then alert person in
charge of TGA.
25. Turn on the air valve on the control board behind the TGA.
26. Set the flow controllers to 60 (for the left controller) and 40 (for the right
controller).
27. Place aluminum sample pan in TGA sample bucket.
28. Push the “Tare” button on the instrument. The TGA micro-balance will
tare with the empty sample pan.
29. Open the program “Thermal Advantage” (TAAdv) on the des top to
design the TGA experiment.
30. Within the Thermal Advantage software choose the following:
a. Setup up the Experiment
i. Change the sample name
ii. Change the filename and folder where the data will be
saved
iii. Choose custom
b. Procedure
i. Choose editor and design the experiment
206
1. Ramp 22°C to 150°C, 3°C/min
2. Isothermal 150°C, 120 minutes
3. Ramp 150°C to 600°C, 3°C/min
c. Notes
i. Change the operator
ii. Pan type is aluminum
iii. Air 60/40
d. Hit Apply
31. Remove sample bucket and pan from sample platform stage.
32. Load 1.5 mg of biosilica into sample pan.
33. Move sample pan into sample bucket.
34. Move sample bucket onto sample platform.
35. Push the “Load” button on the TGA instrument. The sample platform will
move and the wire will pick up the sample bucket.
36. Once the sample is loaded, push the green arrow on the Thermal
Advantage software.
37. The TGA will run for 5.2 hours. When the experiment is complete the
sample platform will exit the furnace and “run complete” will appear on
the Thermal Advantage software.
38. To view the TGA data, open the program “Universal Analysis”.
39. Open the TGA file under File, then Open File.
40. To export into excel, close all open excel files. Choose View, Data Table,
Spreadsheet and All Data Points. Excel will automatically open and the
file can be saved.
41. Turn off the compressed air at the tank.
42. Turn off the air valve on the distribution board behind the TGA.
43. Do not turn off the TGA.
44. Do not exit out of the Thermal Advantage or Universal Analysis Programs.
Additional Instrument Specifications
Heating Rate Range: 0.1 to 100°C/min
Temperature Range: 25°C to 600°C with aluminum sample pan
Weighing Capacity: 1.0 g
Balance Measurement Resolution: 0.10 µg
The TGA balance is sensitive to ambient room temperature and mechanical
vibrations.
207
Procedure A-26.
Thermally Grown SiO2 Preparation for PL Sensitivity Study
Debra Gale (revised June 22, 2008)
Specialized Equipment
 Furnace
 Clean Silicon wafer
 Glass cutter
 Tweezers
Reagents
 Hydrogen peroxide, 30 wt. %
 Methanol
 Deionized water
Procedure
1. Cut the silicon wafer into square coupons 5 mm by 5 mm in dimension.
2. Clean the silicon wafer by treatment with hydrogen peroxide. Use the
same procedure to extract silica frustules from living cells with
hydrogen peroxide to clean the silicon wafer coupons.
3. Clean the Si wafer coupons with deionized water and methanol after
cleaning with hydrogen peroxide.
4. Place the wafer chips into the preheated furnace at 1000ºC for 3 hours.
5. Using tweezers remove the wafer chips and allow to cool to room
temperature.
208
Download