FUNDAMENTAL INVESTIGATION OF INKJET
DEPOSITION AND PHYSICAL IMMOBILIZATION
OF HORSERADISH PEROXIDASE ON
CELLULOSIC SUBSTRATES
by
Sabina Nélida Di Risio
A thesis submitted in conformity with the requirements for the degree of
Doctor of Philosophy
Graduate Department of Chemical Engineering and Applied Chemistry
University of Toronto
© Copyright by Sabina Nélida Di Risio (2009)
- ii -
FUNDAMENTAL INVESTIGATION OF INKJET DEPOSITION
AND PHYSICAL IMMOBILIZATION OF HORSERADISH
PEROXIDASE ON CELLULOSIC SUBSTRATES
Sabina Nélida Di Risio
Doctor of Philosophy
Graduate Department of Chemical Engineering and Applied Chemistry
University of Toronto
2009
ABSTRACT
In this study, novel bio-inks formulated with horseradish peroxidase, HRP, and some
additives were successfully developed for piezoelectric inkjet application. The optimized
bio-ink formulation had a reliable jetting performance and maintained the
biofunctionality before and after printing. The bio-ink also demonstrated a good storage
life for up to 40 days at 4 oC with a negligible loss of biofunctionality. However, it was
observed that some additives used in the bio-ink for obtaining necessary operational
characteristics had detrimental effects on the enzyme activity. Especially, it was found
that various viscosity modifiers typically used in commercial inkjet inks significantly
impaired HRP activity prior to printing. Sodium Carboxymethyl Cellulose was shown to
be an effective viscosity modifier that had no adverse effect on the biological activity of
the HRP enzyme.
Using a confocal scanning fluorescent microscope, a method for characterizing the spatial
distribution of the active enzyme within the cellulosic paper substrates after inkjet
- iii -
printing was developed. Interestingly, it was found that the active printed HRP enzyme
was mostly located in the cell walls of the cellulosic fibers instead of near the pigments or
fillers.
In an effort to better understand the fundamental interactions between the enzyme and the
immobilization substrates, HRP adsorption isotherms on various substrate surfaces were
obtained using the depletion method. The substrates included not only pulp fibers with
varying degree of hydrophobicity and pigment and latex (the key materials used in
papermaking), but also other types of cellulosic fibers of different morphologies,
crystallinities, porosities, or surface charge densities. The influence on enzyme
adsorption and inactivation behaviour of these substrates was compared with that of
polystyrene beads (dialysed), which has been well studied in the literature. Results from
this thesis indicated that hydrophobic interactions between the enzyme and the substrate
surfaces had a major impact on the HRP adsorption behavior, while electrostatic
interactions played a minor role. However, strong hydrophobic interactions could lead to
enzyme inactivation. Research findings from this study suggested that cellulosic pulp
fibers could be tailor-made into excellent enzyme immobilization supports by using
existing fiber surface modification techniques.
- iv -
ACKNOWLEDGEMENTS
This thesis represents the completion of a journey and the beginning of a new one.
Many people and institutions helped me make this journey and my sincere gratitude is
with them.
I am grateful and indebted to Professor Ning Yan, my advisor, for her outstanding
mentorship. With encouragement, challenges, continuous support and patience Prof. Yan
helped me grow as a scientist and opened the doors for me to exciting research fields. I
have been privileged with her wise guidance, trust and generosity.
I would like to thank the members of my examination committee, Professor Mark
Kortschot, Professor Brad Saville and Professor Ramin Farnood, for the stimulating
discussions and their commitment to my professional development. I also would like to
thank Prof. Edgar Acosta, Prof. Mohini Sain and Prof. Per Claesson, members of my oral
defense examination committees, for their interest in my research and valuable feedback.
The technical assistance of Illya Gourevich, Peter Brodersen, Matthew McDonald,
Andrew Lee, Chong Liang, Kieron Moore, Tony Ung, Candida DaCosta, Syed Abthagir
and Carlos Quijano Solis, in different steps of my experimental work is gratefully
acknowledged. I am thankful to Gordon Sisler and Gail Song from the Xerox Research
Centre of Canada for facilitating the use of their instrument for contact angle
measurements.
The financial support from Sentinel, the NSERC Bioactive Paper Network, and
the University of Toronto is gratefully acknowledged. Being part of Sentinel was an
excellent and rich learning experience and my gratitude is with Professors Robert Pelton,
Christopher Hall, Richard Kerekes, Theo van de Ven and Dr. George Rosenberg for their
vision, leadership, and enthusiastic feedback.
- v -
I am also thankful to Hercules, Omya, BASF, Cordenka and Tembec for
providing samples necessary for this study and to Professors Acosta, Farnood, Kortschot,
Woodhouse, Jia and Chan for facilitating the use of their instruments and labs.
Special thanks go to Professor Douglas Reeve and the staff (Joan, Jacquie, Joan
Chen, Paul, Pauline, Trisha, Gorette, Leticia, Julie, Kathy) at the Department of Chemical
Engineering and Applied Chemistry, as well as Professor Honghi Tran and the staff
(Cindy and Anna) at the Pulp and Paper Center in the University of Toronto for making
my journey pleasant and inspirational.
I am grateful to both my old and my new friends, Eugenia, Humberto, Patricia,
Ingrid, Flor, Olive, Daniel, Carlos, Kevin, Erica, Lorena, and colleagues from labs ES
2008 and WB 419, for generously sharing their time, discussions, prayers, and their
enthusiasms with me.
Esta tesis esta dedicada a mi familia por sus incansables desvelos, su amor
incondicional y soporte continuo durante este y todos mis emprendimientos en la vida.
Esta tesis esta también dedicada a Ernesto, mi compañero en este viaje.
¡Gracias Señor por todas tus bendiciones!
- vi -
TABLE OF CONTENTS
ABSTRACT ii
ACKNOWLEDGEMENTS iv
TABLE OF CONTENTS vi
LIST OF TABLES xv
LIST OF FIGURES xvii
LIST OF SYMBOLS AND ABREVIATIONS xxvii
LIST OF SUPLEMENTARY MATERIAL xxxii
Chapter 1 - INTRODUCTION
1.1 - MOTIVATION AND SIGNIFICANCE
1.2 - SCOPE
1.3 - HYPOTHESES
1.4 - OBJECTIVES
1.5 - THESIS OVERVIEW
1
2
3
3
5
Chapter 2 - LITERATURE REVIEW
2.1 - INTRODUCTION TO BIOACTIVE PAPERS
2.2 - ENGINEERING BIOACTIVE PAPERS
2.2.1 - The function: bioanalysis
2.2.2 - The bioagent: a model enzyme
2.2.2.1 - Biomolecules
7
9
9
15
15
- vii -
2.2.2.2 - Horseradish peroxidase
2.2.3 - The solid support: cellulose-based fibrous materials
2.2.4 - The deposition system: inkjet technology
2.2.4.1 - Technological options for high speed manufacturing of
bioactive papers
2.2.4.2- Bioprinting: printing bioagents on paper
(a) - Contact Dispensing
(a1) - Gravure
(a2) - Flexography
(a3) - Screen printing
(a4) - Microspotting or pin printing
(a5) - Microstamping or softlithography
(b) - Non contact dispensing
(b1) - Photolitography
(b2) - Electrospray deposition (ESD)
(b3) - Biological laser printing (BioLPTM)
(b4) - Continuous flow microfluidic printing (CFM)
(b5) - Inkjet printing
b.5.1 - Principle of operation, benefits and challenges
b.5.2 - Piezoelectric jetting parameters: jetting cycle and
maintenance operations
b.5.3 - Jetting performance criteria: feasibility and
reliability
b.5.4 - Non conventional applications of inkjet
technology
17
19
21
21
22
22
22
23
23
24
24
25
25
26
28
29
29
29
31
33
34
- viii -
b5.5 - Enzyme inkjet printing
2.2.5 - Paper-biomolecule attachment: passive adsorption
2.2.5.1 - Immobilization strategies
(a) - Adsorption
(b) - Entrapment
(c) - Confinement
(d) - Cross-linking
(e) Covalent binding
(f) - Bioaffinity
2.2.5.2 - Paper-enzyme interactions during adsorption
2.2.6 - Detection system: colorimetric
2.3 - EVALUATING BIOACTIVE PAPER PERFORMANCE
2.4 - CONCLUSIONS
36
37
37
40
40
41
41
42
42
43
43
45
47
Chapter 3 - EXPERIMENTAL APPROACH
3.1 - BIO-INK FORMULATION
3.1.1 - Enzyme
3.1.2 - Chromogenic enzyme substrate
3.1.3 - Buffer
3.1.4 - Additives
3.1.5 - Liquid vehicles formulations
3.1.6 - Standard bio-ink formulation
3.1.7 - Viscosity measurements
3.1.8 - Surface tension measurements
3.2 - SOLID SUPPORTS
49
49
49
49
49
50
50
51
51
51
- ix -
3.2.1- Fibrous supports for printing
3.2.1.1 - Commercial papers
3.2.1.2 – Handsheets with increasing hydrophobicity
3.2.1.3 - Coating layer
3.2.2 -Supports for adsorption
3.2.2.1 – Model sorbents
3.2.2.2- Treated fibers
3.2.2.2.1- Beating
3.2.2.2.2 - Internal sizing
3.2.2.2.3 - TEMPO-mediated oxidation of ground rayon and
BKSW fiber
3.2.2.2.4 – Surface charge modification using polyelectrolyte
3.2.2.3 – Characterization of model sorbents and treated fibers
3.2.2.3.1 – Surface energy of model sorbents
3.2.2.3.2 – Surface energy of treated fibers
3.2.2.3.3 – Zeta Potential of model sorbents and fibers
3.2.2.3.4 – Scanning electron microscopy (SEM) of model
sorbents and fibers
3.2.2.3.5 – Specific surface area of model sorbents and fibers
3.2.2.3.6 – Carboxylate content of oxidized cellulosic
sorbents
3.2.2.3.7 – X-ray photoelectron spectroscopy of the
handsheets
3.3 - BIO-INK PRINTING
3.3.1-Inkjet printer
51
51
52
53
53
53
54
54
54
54
55
55
55
57
59
59
59
60
61
62
62
- x -
3.3.2- Ink-material compatibility
3.3.3- Control of jetting performance
3.3.4- Jettability test
3.3.5-Printed patterns
3.4 - ENZYME MEASUREMENTS
3.4.1- Activity in solution
3.4.2- Activity in solution (after printing)
3.4.3 - Protein concentration in solution
3.5 - SOLID PHASE BIOANALYSIS OF H2O2
3.5.1-Principle
3.5.2-Color development
3.5.3-Measurement of color response
3.5.4-Bioanalysis calibration
3.6 - ENZYME SPATIAL DISTRIBUTION
3.6.1 - Principle
3.6.2 - New bio-ink formulation
3.6.3 - Inkjet printing and fluorescence development
3.6.4 - Embedding and sectioning
3.6.5 - Confocal laser scanning microscopy (CLSM)
3.7 - ADSORPTION ISOTHERMS
3.7.1 - Enzyme solutions
3.7.2 - Depletion method
3.7.3 - Adsorption isotherm construction
3.7.4 - Inactivation isotherm construction
63
64
64
66
66
66
67
67
68
68
69
69
70
71
71
71
71
72
72
72
72
73
73
74
- xi -
3.7.5 - Modeling of adsorption isotherms
3.7.5.1 - Langmuir’s modeling
3.7.5.2 - Freundlich’s modeling
3.8 - STABILITY MEASUREMENTS
3.8.1 - Bio-ink storage stability
3.8.2 - Bioactive paper storage stability
3.8.3-Adsorbed HRP thermal stability
74
74
75
75
75
75
76
Chapter 4 - INK FORMULATION AND PIEZOELECTRIC INKJET
PRINTING
4.1 - INTRODUCTION
4.2 - SPECIFIC OBJECTIVE
4.3 - RESULTS
4.3.1 - Control of bio-ink surface tension
4.3.2 - Control of bio-ink viscosity
4.3.3 - Bio-ink storage stability
4.3.4 - Printing window for HRP bio-ink
4.3.5 - First drop problem
4.3.6 - Impact of jetting on HRP activity
4.3.7 - Bioactive paper storage stability
4.4 - CONCLUSIONS
77
79
79
79
81
85
86
89
90
91
95
Chapter 5 – FIBROUS MATERIALS AS SUPPORTS FOR
BIOACTIVE PAPERS
5.1 - INTRODUCTION
5.2 - SPECIFIC OBJECTIVE
96
96
- xii -
5.3 - RESULTS
5.3.1 - Bioanalysis of H2O2 on fibrous substrates
5.3.1.1 - Color profile
5.3.1.2 - Bioanalytical performance
5.3.2 - HRP cross-sectional distribution in commercial papers
5.3.3 - HRP cross-sectional distribution in coating layers
5.3.4 - HRP cross sectional distribution in handsheets with an
increasing degree of sizing
5.4 - CONCLUSIONS
97
97
97
99
103
106
107
110
Chapter 6 - PAPER-ENZYME INTERACTIONS
6.1 - INTRODUCTION
6.2 - SPECIFIC OBJECTIVE
6.3 - RESULTS
6.3.1 - HRP adsorption and activity on model sorbents
6.3.2 - Impact of the type of surface charge on HRP adsorption and
activity
6.3.3 - Impact of the surface charge density on HRP adsorption and
activity
6.3.4 - Impact of internal sizing on HRP adsorption and activity
6.3.5 - Modeling of HRP adsorption on cellulosic fibers
6.3.6 - Surface characterization of internally sized cellulosic fibers
by X-ray photoelectron spectroscopy (XPS)
6.3.7 - Thermal stability of HRP adsorbed on cellulosic fibers
6.4 – CONCLUSIONS
111
112
112
112
117
120
123
126
130
133
135
- xiii -
Chapter 7 - CONCLUDING REMARKS
7.1 – CONTRIBUTIONS
7.2 – SPECIFIC CONCLUSIONS
7.2.1 - Ink formulation and piezoelectric inkjet printing of
horseradish peroxidase
7.2.1.1 - Effect of bio-ink additives on enzyme activity
7.2.1.2 - Effect of the jetting process on enzyme activity
7.2.2 - Fibrous materials as support for bioactive papers
7.2.2.1 - Effect of the paper support on the bioanalytical
performance of printed HRP
7.2.2.3 - Effect of the paper support on the spatial distribution
and activity of printed HRP
7.2.3 - Paper-enzyme interactions
7.2.3.1- Fundamental study of the impact of cellulosic
immobilization supports on the adsorption behaviour of HRP
7.2.3.2 - Impact of hydrophobicity on HRP adsorption behavior
on model supports
7.2.3.3 - Impact of surface charge sign and density on HRP
adsorption behavior on model supports and fibers
7.2.3.4 – Impact of internal sizing on HRP adsorption behavior
on cellulosic fibers
7.3 - RECOMMENDATIONS
138
138
139
139
139
140
140
141
142
142
143
143
144
145
REFERENCES
147
- xiv -
APPENDIX
A.1 - Enzymatic Assay of HRP with ABTS as substrate I
A.2 - Temporal evolution of color response of HRP-printed uncoated
woodfree paper
III
A.3 - Fluorescent spectral response of Amplex Red and paper IV
A.4 - Storage stability for HRP enzyme (Sigma P2088) as reported by the
supplier website
V
A.5 - Impact of temperature on enzyme activity measurements VI
A.6 - Driving waveform for bio-ink piezoelectric inkjet printing VII
A.7 - Diffuse reflectance spectra of colored spots on different commercial
paper substrates
VIII
A.8 - Bioanalytical performance of commercial papers IX
A.9 - Degree of sizing for commercial uncoated papers XI
A.10 - Degree of sizing for handsheets XII
A.11 - Methylene blue adsorption isotherms on model sorbents and fibers XIII
A.12 - Conductometric titrations of untreated and oxidized sorbents XIV
A.13 - Non Linear Regression Outputs from Sigmaplot XVI
A.14 - Chemical structure and proposed sizing mechanism for AKD and
rosin-alum systems
XXX
- xv -
LIST OF TABLES
2.1 Range of affinity constants for some typical biomolecules.
(Adapted from [7])
11
2.2 Considerations related to the design of an immobilization
support. (Adapted from [36] and [13])
20
2.3 Non conventional applications of the inkjet technology 35
2.4 Previous work on inkjet printing of enzymes 36
2.5 Immobilization objectives [130,133] 38
3.1 Liquid vehicles formulations 50
3.2 Commercial paper supports 52
3.3 Surface energy of model sorbents as reported in the
literature
56
3.4 Surface energy components of test liquids 58
3.5 Methods used for evaluation of specific surface area 59
3.6 Laser confocal microscopy conditions 72
4.1 Comparison between a generic commercial ink and
enzyme bio-ink formulations
78
4.2 Jetting conditions
88
- xvi -
5.1 Color response developed by different papers printed with
standard bio-ink after exposure to a 2mM H2O2 solution.
Paper codes: (A) chromatographic paper, (B) uncoated
mechanical paper, (C) uncoated recycled paper, (D)
uncoated wood-free paper, (E) color copy cover, (F) coated
grade for offset, (G) coated grade for inkjet, and (H) cast
coated paper.
98
6.1 Properties of the model sorbents 113
6.2 Properties of the rayon and fibers with and without
oxidation treatment
120
6.3 Properties of the fibers 124
6.4 Binding affinity constants, K, obtained by Langmuir’s fit
of the experimental HRP adsorption isotherms from Figure
6.6
129
6.5 Heterogeneity index, m, obtained by Freundlich’s fitting of
the experimental HRP adsorption isotherms from Figure
6.6
130
6.6 XPS analysis of unsized and increasingly sized handsheets 132
- xvii -
LIST OF FIGURES
1.1 Thesis overview and organization of the chapters 6
2.1 Simplified scheme of bioactive paper principle for
bioanalysis
8
2.2 Potential areas of application for bioactive papers 9
2.3 Range of analyte concentrations measured by different
bioanalytical systems. (Adapted from [9])
13
2.4 Relative sizes for biomolecules and fibers 16
2.5 The catalytic cycle of HRP C with a generic reducing
substrate
18
2.6 Potential technologies for mass production of bioactive
papers
21
2.7 Gravure 22
2.8 Flexography 23
2.9 Screen printing 23
2.10 Microspotting 24
2.11 Microcontact printing 25
2.12 Photolithography 26
2.13(a) Distal ESD 27
2.13(b) Proximal ESD 27
- xviii -
2.14 Biological laser printing. (Adapted from [69]) 28
2.15 Continuous- flow microfluidic printing 29
2.16 Main inkjet technologies. (Adapted from [75]) 30
2.17 Cross section of a single piezoelectric drop-on-demand
ejector
31
2.18 Generic piezoelectric jetting cycle 32
2.19 Immobilization strategies. (Some schemes were adapted
from [134])
39
2.20 Detection systems 44
3.1(a) Inkjet material deposition system 62
3.1(b) Cartridge and print-head 62
3.1(c) Printhead nozzles: 16 nozzles, 254µm spacing, 21.5µm
diameter.
63
3.2 Measurement and control of jetting performance 64
3.3 Sketch (not to scale) of inkjet printed patterns. Spot (left),
lines (center) and dots (right)
66
3.4 Schematic illustration of H2O2 bioanalysis using paper 69
4.1 Effect of Triton X-100 dosage on surface tension of 40mM
potassium phosphate buffer at pH 6.8 – The error bars
represent the ± one standard deviation for 6 replicate
surface tension measurements of the same sample. The
maximum observed value of the standard deviation was
0.8mN/m.
80
- xix -
4.2 Effect of Triton X-100 dosage on activity of 0.05µM HRP
in 40mM potassium phosphate buffer at pH 6.8. Three
replicate activity measurements of both the sample and the
control were performed. The change in activity was
calculated with the average activity values corresponding to
the sample and the control. Variations of ±10% are within
the experimental error of the HRP activity assay (See
section 3.4.1).
81
4.3 Effect of the dose of different viscosity modifiers on the
ambient temperature viscosity of solutions containing
40mM potassium phosphate buffer pH 6.8 and 0.1 wt.-%
Triton X-100. The error bars represent the ± one standard
deviation for 3 replicate viscosity measurements of the
same sample. The maximum observed value of the standard
deviation was 0.2 cps (corresponding to 0.01 in logarithmic
viscosity scale).
82
4.4 Impact of viscosity on HRP activity for a solution
containing 0.05 µM HRP in 40mM potassium phosphate
buffer at pH 6.8 with 0.1 wt.-% Triton X-100 and variable
doses of the viscosity modifiers indicated in the plot legend.
Three replicate activity measurements of both the sample
and the control were performed. The change in activity was
calculated with the average activity values corresponding to
the sample and the control. Variations of ±10% are within
the experimental error of the HRP activity assay (See
section 3.4.1).
83
- xx -
4.5 Temporal evolution of peroxidase activity in the standard
HRP bio-ink. Control: 50U/ml HRP in PBS pH6.8; Sample:
50U/ml HRP in 40mM potassium phosphate buffer at
pH6.8 with 0.1wt.-% Triton X-100, 0.5wt.-% CMC, and
10wt.-% glycerol. Activity measured in terms of change in
absorbance per unit time. Error bars represent the ±one
standard deviation corresponding to three replicate
measurements of the same sample.
86
4.6 Velocity profiles of 0.05µM HRP in 40mM potassium
phosphate buffer pH6.8, 0.1wt.-% Triton X-100 and
variable amounts of CMC. Firing frequency is 3 KHz,
driving voltage is 30V and ink firing temperature is 28°C
87
4.7 Driving waveform for reliable jetting of HRP bio-ink 89
4.8 Comparison of enzyme activities in solution for different
steps during printing. Error bars represent the ±one
standard deviation of four independent ink samples. Each
sample activity corresponds to the average of three
replicate measurements. Results of Student’s two-tail T-test
indicated on the plot.
91
4.9 Temporal evolution of the color response (measured as AR)
of HRP-printed paper stored in a freezer (-20°C) and
exposed to solutions of low, intermediate, and high H2O2
concentrations. Error bars represent the ±one standard
deviation in the AR value corresponding to the simultaneous
color development of ten enzyme spots printed on the same
paper strip.
92
- xxi -
4.10 Temporal evolution of the color response (measured as AR)
of HRP-printed paper stored in a fridge (4°C) and exposed
to low, intermediate, and high concentration of H2O2. Error
bars represent the ±one standard deviation in the AR value
corresponding to the simultaneous color development of ten
enzyme spots printed on the same paper strip.
93
4.11 Temporal evolution of the color response (measured as AR)
of HRP-printed paper stored in a conditioned room for
paper testing (23°C, 50%RH) and exposed to low,
intermediate, and high concentration of H2O2. Error bars
represent the ±one standard deviation in the AR value
corresponding to the simultaneous color development of ten
enzyme spots printed on the same paper strip.
94
5.1 Calibration curve of AR vs. log C for H2O2 bioanalysis
performed on uncoated wood free paper printed with HRP
bio-ink. C = log (109. [H2O2]) mol.l-1. Error bars represent
the ±one standard deviation in the AR value corresponding
to the simultaneous color development of ten enzyme spots
printed on the same paper strip.
100
5.2 Sensitivity for H2O2 solid phase bioanalysis on different
commercial paper supports as detailed in Table 3.2. Paper
codes: (C) uncoated recycled paper, (D) uncoated wood-
free paper, (E) color copy cover, (F) coated grade for offset,
and (G) coated grade for inkjet.
101
- xxii -
5.3 Limits of detection and linear range of measurement for
H2O2 solid phase bioanalysis on different commercial paper
supports as detailed in Table 3.2. Open and closed squares
correspond to lower and upper limits of detection,
respectively. Paper codes: (C) uncoated recycled paper, (D)
uncoated wood-free paper, (E) color copy cover, (F) coated
grade for offset, and (G) coated grade for inkjet.
102
5.4 Combined cross-sectional CLSM images of the active
HRP enzyme (red) and pigments/fillers (green) for
negative controls, inkjet printed sample, and positive
control of commercial papers. Paper codes: (C) uncoated
recycled paper, (D) uncoated wood-free paper, (E) color
copy cover, (F) coated grade for offset, and (G) coated
grade for inkjet. For paper E only partial cross-sectional
view is shown. Bar = 50µm.
104
5.5 CLSM cross-sectional images of a thick coating layer with
HRP enzyme printed on the left side and exposed to H2O2
solution on the right side. Left: pigment map. Center: active
HRP enzyme map. Right: overlay of pigment and enzyme
maps.
106
5.6 Combined surface view CLSM images of the active HRP
enzyme (red) and fibers (green) for handsheets increasingly
sized from 0 to 1.6wt-% of a rosin based sizing agent. Bar =
400µm.
108
5.7 Combined cross-sectional CLSM images of the active HRP
enzyme (red) and fibers (green) for handsheets increasingly
sized from 0.8 to 1.6wt-%. Bar = 50µm.
109
- xxiii -
5.8 Combined surface view CLSM images of the active HRP
enzyme (red) and fibers (green) for handsheets increasingly
sized from 0 to 0.3wt-% with AKD based sizing agent.
Bar= 400µm.
109
6.1 SEM micrographs of model sorbents a) Microcrystalline
cellulose (400X), b) Ground rayon filament yarn (200X), c)
Ground calcium carbonate (4500X), d) Polystyrene beads
(20000X), and e) SB latex (3000X).
114
6.2 Adsorption isotherms of HRP on model surfaces. Error bars
represent the ± one standard deviation for two
independent adsorption experiments. Protein measurements
for each adsorption experiment represent an average of two
sample measurements.
115
6.3 Inactivation isotherms of HRP on model surfaces. The
±20% area is the uncertainty range for the inactivation
measurements. Error bars represent the ± one standard
deviation for two independent adsorption experiments.
Protein and activity measurements for each adsorption
experiment represent an average of two sample
measurements.
117
6.4 Adsorption isotherms of HRP on rayon with positive and
negative surface charges. Error bars represent the ± one
standard deviation for two independent adsorption
experiments. Protein measurements for each adsorption
experiment represent an average of two sample
measurements.
118
- xxiv -
6.5 Inactivation isotherms of HRP on rayon with positive and
negative surface charges. The ±20% area is the uncertainty
range for the inactivation measurements. Error bars
represent the ± one standard deviation for two independent
adsorption experiments. Protein and activity measurements
for each adsorption experiment represent an average of two
sample measurements.
119
6.6 Adsorption isotherms of HRP on rayon and cellulosic fibers
with and without TEMPO-mediated oxidation treatment.
Error bars represent the ± one standard deviation for two
independent adsorption experiments. Protein measurements
for each adsorption experiment represent an average of two
sample measurements.
121
6.7 Inactivation isotherms of HRP on rayon and cellulosic
fibers with and without TEMPO-mediated oxidation
treatment. The cellulosic fibers with and without TEMPO-
mediated oxidation treatment. Error bars represent the ±
one standard deviation for two independent adsorption
experiments. Protein and activity measurements for each
adsorption experiment represent an average of two sample
measurements.
122
6.8 Adsorption isotherms of HRP on beaten bleached kraft
softwood fibers with an increasing degree of internal sizing
(hydrophobicity). Error bars represent the ± one standard
deviation for two independent adsorption experiments.
Protein measurements for each adsorption experiment
represent an average of two sample measurements.
125
- xxv -
6.9 Inactivation isotherms of HRP on beaten bleached kraft
softwood fibers model surfaces with an increasing degree of
internal sizing (hydrophobicity). The ±20% area is the
uncertainty range for the inactivation measurements. Error
bars represent the ± one standard deviation for two
independent adsorption experiments. Protein and activity
measurements for each adsorption experiment represent an
average of two sample measurements.
126
6.10 Experimental [○] HRP adsorption isotherms on increasingly
sized cellulosic fibers fitted with Langmuir’s [ ̶ ] and
Freundlich’s [...] empirical models. A. Untreated, B. 0.8wt-
% rosin-sized C. 1.6wt-% rosin-sized, D. 0.3wt-% AKD-
sized.
128
6.11 X-Ray Photoelectron C1s spectra for increasingly sized
handsheets. The XPS spectra were not smoothened prior to
deconvolution. A Gaussian-Lorentzian ratio of 70%/30%
was used for peak deconvolution. The binding energy scale
was referenced to the C1s line of aliphatic carbon set at
285.0 eV
131
6.12 DSC control thermograms for enzyme. A) 40mM KH2PO4
buffer pH 6.8, B) 10mg/ml HRP solution in buffer, C)
Same as B after dialysis against buffer.
134
- xxvi -
6.13 DSC thermograms for wet fibers after 24h of enzyme
adsorption from 0.4mg/ml HRP solutions followed by 30
min centrifugation at 10,000 rpm to remove supernatant.
A) Blank: buffer adsorbed on untreated fiber, B) HRP
adsorbed on untreated fiber, C) HRP adsorbed on fibers
treated with 0.8wt-% rosin-based sizing, D) HRP adsorbed
on fiber treated with 1.6wt-% rosin-based sizing, E) HRP
adsorbed on fiber treated with 0.3wt-% AKD sizing.
135
- xxvii -
LIST OF SYMBOLS AND ABREVIATIONS
ABTS 2, 2'-azino-di-(3- ethylbenzthiazoline) 6-sulphonate
AC Amorphous cellulose
AKD Alkyl ketene dimer
AH2 Generic hydrogen donor
Amplex Red 10-acetyl-3,7-dihydroxyphenoxazine
Ar Argon
αCP Affinity contact printing
BET Brunauer, Emmett and Teller theory
BioLPTM Biological laser printing
BKSW Bleached kraft softwood
BSA Bovine serum albumin
C1 Aliphatic primary carbon
C2 Secondary carbon (alcohol, ether)
C3 Tertiary carbon (acetal, ketone, aldehyde)
C4 High binding energy carbon (ester, carboxylic acid)
CCD Charged-couple device
CFM Continuous flow microfluidic
CIJ Continuous inkjet
- xxviii -
CLSM Confocal Laser Scanning Microscopy
cmc Critical micelle concentration
CMC Sodium carboxymethyl cellulose
COOH Carboxylic acid
CP Contact processing
CTP Centre Technique de l’Industrie des Papiers, Cartons et
Celluloses de Grenoble
D65 Daylight standard illuminant defined by the International
Commission on Illumination (CIE)
DCA Dynamic contact angle
DMSO Dimethylsulfoxide
DNA Deoxyribonucleic acid
DOD Drop-on-demand
d.p.i. Dots per inch
DSC Differential Scanning Calorimetry
EG Ethylene glycol
EPDM Ethylene propylene diene M-class rubber
ESD Electrospray deposition
FQA Fiber quality analyzer
GAL Galactosidase enzyme
- xxix -
GCC Ground calcium carbonate
GOD Glucose oxidase enzyme
GvOC Good-van Oss-Chaudhury
HCl Hydrochloric acid
HeNe Helium-neon laser
His-Tag Hexa-histidine tag
HRP Horseradish peroxidase enzyme
HRP A Acidic horseradish peroxidase isoenzyme
HRP B Neutral horseradish peroxidase isoenzyme
HRP C Neutral horseradish peroxidase isoenzyme
HRP D Basic horseradish peroxidase isoenzyme
HRP E Basic horseradish peroxidase isoenzyme
HST Hercules Sizing Test
H2O2 Hydrogen peroxide
IGC Inverse gas chromatography
ISFET Ion selective field effect transistor
KH2PO4 Potassium phosphate monobasic
LED Light-emitting diode
LLD Lower limit of detection
- xxx -
LOD Lactate oxidase
LWC Light weight coating
MCC Microcrystalline cellulose
MEMS Micro-electro-mechanical system
MWCO Molecular weight cut off
µCP Micro-contact printing
N2 Nitrogen gas
NaCl Sodium chloride
NaOH Sodium hydroxide
PAH Poly-(allylaminehydrochloride)
PCC Precipitated calcium carbonate
PEDOT-PSS Poly- (3,4-ethylenedioxythiophene/polystyrene sulfonic acid)
PDMS Poly-(dimethyl siloxan)
PEG Poly-(ethylene glycol)
PFI Paper and Fibre Research Institute (Norway)
PS Poly-(styrene)
PVA Poly-(vinyl alcohol)
RH Relative humidity
SAM Self assembled monolayer
- xxxi -
SB Styrene butadiene latex
SBA Styrene butadiene acrylonitrile latex
SEM Scanning electron microscopy
SSA Specific surface area
TAPPI Technical Association of the Pulp and Paper Industry (USA)
TEMPO 2,2,6,6-tetramethyl-1-piperidininyloxy radical
Triton Iso-octyl phenol monoethylene glycol ether
TOF-SIMS Time-of-Flight Secondary Ion Mass Spectroscopy
UV Ultraviolet
UV-VIS Ultraviolet-Visible
VTT Technical Research Centre of Finland
XPS X-ray Photoelectron Spectroscopy
- xxxii -
LIST OF SUPPLEMENTARY MATERIAL
4.1 Movie of the drop firing process for the bio-ink containing
0.5wt-% CMC with optimized jetting conditions
File name: CMC05.avi
89
- 1 -
Chapter 1
INTRODUCTION
1.1-MOTIVATION AND SIGNIFICANCE
There is a growing interest in developing bioactive paper products that can
incorporate advanced biological functionalities as inexpensive devices to detect,
capture or inactivate analytes (including pathogens) in water, food, air or human
fluids. Diagnostic papers, pathogen trapping papers, smart packaging and
security papers are just some examples of the many possible applications.
Despite of numerous advantages that the bioactive papers may offer, there are
currently very few successful cases of commercial application of the concept.
To fabricate low cost bioactive paper products, a suitable high-speed mass-
production manufacturing process is necessary. Papermaking (web forming,
coating and surface treatment) and printing (contact and non contact) processes
are some of the potential high-speed high-volume manufacturing methods for
making these novel bioactive paper products economically. Significant waste of
costly biomolecules during some papermaking unit operations and high risk of
biomolecule inactivation and denaturing under harsh operating conditions (pH,
additives, temperature, mechanical action, dry conditions) are some of the
drawbacks associated with the papermaking process. On the other hand, printing
technologies appear to be better suited for delivering biomolecules onto
surfaces. Particularly, non contact printing methods provide a cleaner deposition
process for the bioagents with a reduced contamination risk and no mechanical
impact. The non contact printing technologies that do not expose the bio-ink to
denaturing conditions will clearly be more advantageous.
Among various non contact printing methods, inkjet printing technology has
some unique benefits in delivering biological solutions onto solid materials.
- 2 -
Some major benefits include small deposition volume, clean non contact
operation, good speed, accurate placement, and high spatial resolution. Although
there are some preliminary feasibility studies of using inkjet printers for
biomolecule deposition in the literature, a more systematic study of bio-ink
formulation (i.e., inks that contain a biomolecule as the main active ingredient)
and inkjet printing condition is still highly necessary in order to develop
optimized bio-inks with good jetting performance.
Paper is mostly made from cellulosic pulp fibers. Cellulose, being one of the
most abundant biopolymers on earth, has many desirable attributes to be
considered as a solid support for bioagent immobilization. It is inexpensive, non
toxic, and biodegradable. It is obtained from renewable resources and has good
mechanical properties. However, cellulose also poses some unique challenges
when used as a solid support due to its inherent structural complexity (at the
molecular, supramolecular and morphological levels).
To develop bioactive papers, it is also important to devise suitable
immobilization methods to place the biomolecules on the paper surfaces while
maintaining biofunctionality. Even in the case of immobilization through simple
physical adsorption, the intricate structure and composition of the biomolecules
combined with the heterogeneity and structural complexity of the cellulosic
paper substrates make the interactions between the biomolecules and the
cellulosic supports highly complicated. Thus, a better understanding of the
fundamental interactions between the biomolecules and the cellulosic substrates
is needed in order to engineer suitable paper supports with optimum structures
and superior performance.
1.2- SCOPE
In this study, horseradish peroxidase (HRP) enzyme was selected as the bioagent for the
investigation. HRP is a well-characterized and robust enzyme that is commercially
available with relatively low cost. Based on the solid phase bioanalysis of H2O2 using
HRP, the scope of the thesis research includes the following areas:
- 3 -
1. Effect of inkjet ink additives on biomolecule activity in bio-ink formulations
2. Impact of the jetting process on the biomolecule functionality
3. Significance of the support in relation to the bioactive paper performance
4. Extent of biomolecule spatial distribution and functionality after immobilization
5. Mechanism of fundamental interactions between the biomolecule and the
cellulosic support
1.3-HYPOTHESES
The following hypotheses were addressed in this thesis:
1. Additives used in piezoelectric-based inkjet ink formulations, specially
surfactants and viscosity modifiers, can lead to inactivation of the HRP
enzyme
2. Suitable jetting parameters can be identified that do not negatively affect
the HRP enzyme activity during the piezoelectric drop-on-demand inkjet
printing process
3. The bioanalytical performance of HRP-printed bioactive papers depends
on the localized activity of the enzyme and is a function of the enzyme
spatial distribution and surface properties of the substrates
4. Optimum cellulosic substrates for bioagent immobilization can be
engineered to have suitable surface energy components and electrostatic
charge characteristics
1.4- OBJECTIVES
The overall objective of this study is to better understand the performance of
paper as an enzyme immobilization support and the efficacy of using inkjet
printing technology for deposition of the bioactive agents. To test these
hypotheses, the following specific set of objectives was identified:
- 4 -
1. To investigate the efficacy of printing HRP on paper substrates using a
piezoelectric drop-on-demand inkjet material deposition system
o To formulate a biologically active and reliably jettable HRP-
containing bio-ink
o To define an operational window for piezoelectric inkjet printing
process of the HRP-containing bio-ink
o To evaluate the impact of the piezoelectric jetting process on the
enzyme activity
2. To relate bioanalytical performance to physical and chemical properties of
the surface of the paper support
o To define indicators for bioanalytical performance of HRP
immobilized on paper
o To identify some paper attributes that can significantly affect
bioanalytical performance
o To propose strategies for controlling and improving the analytical
performance of the bioactive papers
3. To characterize the distribution and uniformity of HRP printed on paper
o To develop a method based on confocal laser scanning
microscopy, CLSM, to image surface and cross-sectional active
HRP enzyme distributions within paper
o To investigate the local distribution of the active enzyme in
different commercial papers
o To investigate the impact of surface treatment of the cellulosic
fibers on the spatial distribution of the inkjet printed HRP
4. To study sorbent-enzyme interactions on solid supports with different surface
charges and surface energy components
o To characterize the adsorption behavior and residual activity of
HRP on the cellulosic fiber surfaces
o To investigate the impact of the fiber treatment on HRP adsorption
behavior and activity
- 5 -
1.5-THESIS OVERVIEW
This thesis contains seven chapters, a list of references, and an appendix.
Immediately following the Introduction chapter, Chapter 2 examines the
conceptual modules required to engineer a bioactive paper for bioanalysis and
provides a pertinent literature review. Unexplored research areas and relevant
knowledge gaps are summarized. Chapter 3 includes a detailed description of
the materials, experimental methods, and equipment used in the study.
Chapters 4 to 7 present the main experimental results and conclusions of this
study. The focus in Chapter 4 is on bio-ink formulation and inkjet deposition.
Chapter 5 explores the impact of different types of paper substrates and
characterizes the local active enzyme distribution using a newly developed
CLSM technique. In Chapter 6, fundamental interactions between the enzyme
and the immobilization support are investigated with a specific emphasis on
hydrophobic and electrostatic interactions. Finally, Chapter 7 summarizes the
main conclusions of this thesis and suggests areas for further study. The main
thesis topics and their interrelationship are outlined in Figure 1.1.
- 6 -
Figure 1.1- Thesis overview and organization of the chapters
Bio-ink Formulation
• Additives• Concentration• Activity• Storage stability
Bio-ink Printing
• Feasibility• Reliability• Activity• Stability after printing
Bio-ink Immobilization
• Spatial distribution• Ink-support interactions• Thermal stability
Bioanalysis
• Performance indicators•Qualitative and quantitative evaluation
- 7 -
Chapter 2
LITERATURE REVIEW
2.1-INTRODUCTION TO BIOACTIVE PAPERS
Bioactive papers have been defined as high value-added fiber-based products
with an advanced biological functionality capable of identifying, capturing
and/or inactivating specific target analytes (e.g. chemical substances, pollutants,
pesticides, toxins, antigens, pathogens, drugs, allergens, etc.) [1,2,3,4]. The
focused development of bioactive paper products is a relatively new initiative
championed by Canadian researchers [5]. Their goal is to develop cheap papers
that can provide early warnings about health or safety risks by means of an
instant visible indication of the presence of harmful substances in water, food or
air without the need of measuring instruments. Although the researchers
anticipated that paper-based biosensing may not be as sensitive as other options,
low cost, portability, ease of use and instant response were thought to be some
clear advantages [2, 3].
The key components of a bioactive paper in bioanalysis are: the bioagent, i.e. the
biological molecule that can specifically recognize the target; the reporter, i.e.
the chemical substance or physical variable that will change when the
biorecognition event occurs; and the support, i.e. the cellulosic fibrous network
where the function is embedded (See Figure 2.1). When target analyte and
bioactive paper are in contact, the embedded biological function is triggered and
the reporter signals the event either qualitatively or both qualitatively and
quantitatively.
- 8 -
a) In the presence of target
b) In the presence of non target
Figure 2.1- Simplified scheme of bioactive paper principle for bioanalysis
There is a wide range of potential uses for bioactive papers; some areas of
application are illustrated in Figure 2.2. Frequently, the application determines
the “function” or “combination of functions” that needs to be incorporated in the
paper, thus, it defines the requirements for the bioagent.
Being a more recent research field little is known about the bioactive papers.
Researchers from VTT Technical Research Centre of Finland published a
comprehensive literature review on the use of biomolecules in functional
materials focusing on early examples and potential applications for bioactive
papers [6]. It provides a good summary of past work and an overview of some
potential areas for innovation.
Fibrous
SupportBioagent Reporter
Bioactive
Paper
Fibrous
SupportTarget
FUNCTION
ON
Fibrous
Support
Non
Target
FUNCTION
OFF
- 9 -
Figure 2.2 – Potential areas of application for bioactive papers
2.2-ENGINEERING BIOACTIVE PAPERS
2.2.1-The function: bioanalysis
Over the last decade, bioanalytical techniques have been increasingly used in
clinical diagnosis, medical research, and pharmaceutical discovery. Also, they
have been progressively replacing traditional chemical analysis in industrial,
defense, food, agricultural, and environmental applications. Moreover,
bioanalytical systems have been used in conjunction with a number of
physicochemical transducers to develop biosensors for quantifying and/or
monitoring single or multiple analytes in complex sample matrices.
Bioactive
Paper
Medical Diagnosis
•Health Monitoring
•Point of Care Biosensors
Environment
•Biohazards
•Smart Buildings
•Environmental Sensors
•Water Purification
•Pesticide Detection
Research
•Life Science Research
•Drug Discovery
•Biofuel cells
Public Health
•Biodefense
•Infectious Diseases
•Epidemics/Pandemics
Smart Packaging
•Food Quality, Freshness and Safety
•Beverages
•Labelling
•Built-in indicators
Pathogen Trapping
•Filters
•Protective Clothing
Security
•Anti-counterfeiting
•Brand Protection
•Document Authentication
- 10 -
Bioanalytical techniques are based on the superior recognition ability of some
biomolecules for specific target analytes. A biorecognition event can be
depicted (see equation 2.1) as the interaction between two binding partners: a
biomolecule (antibody, enzyme, receptor, etc) and a target (antigen, substrate,
hormone, etc), in producing an affinity pair [7].
BBBB + + + + L L L L ↔↔↔↔ BBBBLLLL [2.1]
KKKK aaaa = = = = kkkk aaaakkkkdddd ==== �BBBBLLLL��BBBB��LLLL� [2.2]
Where, B ligate (a biomolecule)
L ligand (target)
BL affinity pair
Ka affinity constant
ka association rate constant
kd dissociation rate constant
The biorecognition event is characterized, at equilibrium, by an affinity constant, Ka,
which describes the strength of the attachment between the binding pair, with higher Ka
values corresponding to stronger binding. Table 2.1 illustrates the order of magnitude for
some typical affinity constants.
- 11 -
Table 2.1 - Range of affinity constants for some typical
biomolecules. (Adapted from [7])
Biomolecule (B) Target (L) Ka (M-1)
Avidin Biotin ~ 1015
Receptor Hormone, toxin, etc 108 - 1012
Antibody Antigen 107 - 1011
Lectin Carbohydrate 103 - 106
Enzyme Substrate 103 - 105
In the particular case of an enzyme, in addition to the biocapturing of the target,
bioconversion takes place [8]. During bioconversion, the enzyme acts as a
highly efficient catalyst in the transformation of the substrate into a product.
Simply described in terms of Briggs-Haldane kinetics, the reversible binding
interaction is followed by an irreversible conversion step,
E E E E + + + + S S S S ��↔��� ES ES ES ES ��→ E E E E + + + + PPPP [2.3]
KKKK mmmm = = = = �kkkk1111++++kkkk ----1111� / / / / kkkk2222 [2.4]
v v v v = = = = VVVVmaxmaxmaxmax ����SSSS� / � / � / � / ����S S S S ����+ + + + KKKKmmmm!!!! [2.5]
- 12 -
Where,
E enzyme
S substrate
ES enzyme-substrate complex
P product
Km Michaelis constant
V rate of reaction
Vmax maximum rate of reaction
k1, k-1, k2 rate constants
Bioanalytical techniques can be grouped into bioassays and biosensors. Bioassays consist
of two separate steps: molecular recognition of an analyte by a biomolecule followed by
detection, i.e. the transformation of the recognition event into a measurable signal (e.g.
color change, electrochemical change, mass change). Depending on the biomolecule
involved, bioassays can be classified into biocatalytic assays (using enzymes) and
bioaffinity assays (using antibodies). Biosensors, on the other hand, ideally incorporate
recognition and detection in one single step allowing rapid and continuous analyte
monitoring. Moreover, biosensors ideally should be reversible i.e. allow biomolecule
reuse over more than one measurement cycle. The range of analyte concentrations that
can be detected with different bioanalytical techniques is illustrated in Figure 2.3.
— 13 —
Figure 2.3 – Range of analyte concentrations measured by different bioanalytical
systems. (Adapted from [9])
It becomes apparent from the previous paragraphs that some advantages of bioanalytical
techniques over chemical analysis are:
• Higher selectivity/specificity, given by the biomolecular recognition event
• Higher sensitivity, given by the very low analyte concentrations that can be measured
• Lower response time, especially in the case of biosensors
• Wider dynamic range, several analyte concentration decades can be measured
• Lower detection limit, a consequence of the higher sensitivity
Other aspects that can favour the use of bioanalytical systems depending on the
nature of the sample are,
• Low cost, when compared to alternate chemical methods relying on expensive
instrumentation
• High throughput capability, several bioassays can be run in parallel (e.g. 96 well-plate
immunoassays)
• Possibility of miniaturization, smaller volumes reduce the expensive biomolecule use
and could provide portability and/or disposability to the sensor
• Multianalysis capability, several analytes can be measured simultaneously
Enzyme Electrodes
Direct Immunosensors
Indirect Immunosensors
Immunoassays
— 14 —
• Simplicity, especially in the case of disposable bioassays that do not require
expensive transducers and/or highly trained analyst to interpret the results of the
analysis
Many bioanalytical systems rely on the creation of a sensitive layer, an interface
between a support and a bioactive material. In order to create that sensitive
layer, a system to deposit the biological solution onto a solid support and a
suitable immobilization strategy are required. While selectivity is an inherent
characteristic of the biomolecule, the sensitivity of the analytical system is
determined by both the biomolecule and the support.
Most biomolecules are proteins. The three-dimensional structure has to be
preserved for proteins to maintain their functionality. Proteins exhibit different
degrees of denaturation if physical conditions such as moisture, temperature,
pressure or mechanical stress, and chemical factors like pH, presence of
denaturing substances (alcohols, heavy metals, detergents) and ionic strength are
not kept at optimum values [10]. Ensuring optimal conditions is not trivial in
bioanalytical systems. Even a slight change in the measurement conditions or in
the sample properties can produce problems ranging from more frequent need of
standardization to complete loss of sensitivity.
Despite the apparent advantages of the bioanalytical methods in terms of
selectivity and sensitivity, a number of difficulties have to be overcome when
bioagents rather than chemical reagents are used. The main challenge appears to
be the stability of the biomolecule itself over time. The support choice, the
deposition system, and the immobilization strategy can all play a significant role
in preserving the functionality of a particular bioagent. Past literature studies on
relevant bioanalytical methods are discussed in detail in the next few sections.
— 15 —
2.2.2 - The bioagent: a model enzyme
2.2.2.1 - Biomolecules
A variety of bioagents have been used before in bioanalytical systems. Amongst
them, antibodies, nucleic acids, and enzymes have received a considerable
amount of attention. As shown in Table 2.1 and Figure 2.3, antibodies exhibit a
high ligate affinity that allows ultrasensitive detection of the antigens; however,
they do not have a catalytic function (except for a special subgroup of antibodies
known as catalytic antibodies). In contrast, enzymes are highly selective
biocatalysts that give a more rapid response. But they are more susceptible to
inactivation. Nucleic acids, as opposed to proteins, are robust molecules which
offer tolerance to a broader range of analysis conditions and they can be easily
amplified and exhibit negligible non specific binding.
The size of the biomolecule is another important consideration, especially for
the selection of the bioprinting technique and immobilization support.. Whole
cells (usually microbial) with sizes in the micrometer range are situated on one
end of the scale, while enzymes and antibodies with nanometer sizes (2 to
100nm [11]) are situated on the other end. Another aspect that needs to be
considered is the possibility of reuse of the biomolecules. Enzymes, due to their
inherent biocatalytic property, participate in the biorecognition event without
being consumed. For other bioagents like antibodies, reuse is problematic
because they have to be regenerated with chemical reagents under conditions
that can impair the antibody functionality.
— 16 —
Figure 2.4 – Relative sizes for biomolecules and fibers
Enzymes and antibodies have found many applications as analytical reagents in
clinical chemistry and in biosensors [12]. Nucleic acids are being widely used in
genomics, while whole cells are being utilized in environmental analysis.
Enzymes are the most frequently used bioagents and are preferred for the
detection of small analytes [8]. In addition, a variety of well characterized
enzymes are available commercially [13].
Printing active proteins (enzymes and antibodies) is more challenging than
printing nucleic acids. Bernard et al. [14] used microcontact printing to produce
protein patterns on solid substrates. They observed that after printing, 100%,
70%, and from 50 to 70% of the original functionality was retained by
antibodies, cells, and enzymes, respectively. Unfortunately, a similar type of
comparison cannot be found for other printing techniques.
1Å
1nm
10nm
100nm
1µm
10µm
100µm
1mm
1cm
1dm
1m
Water
Enzyme
HumansBacterium
Cellulosic
Fiber WidthCellulosic
Fiber Length
Antibody
Dye
DNA
diameter
Virus
Animal cellPaper
Thickness
— 17 —
2.2.2.2- Horseradish peroxidase
An enzyme, horseradish peroxidase (HRP), has been chosen as the model bioagent for
this research. HRP belongs to the group of oxidoreductases (EC 1.11.1.7) with substrate
specificity for hydrogen peroxide [15]. At least fifteen different isoenzymes have been
isolated from horseradish roots (amoracia rusticana) and a great number of isoforms
have been detected by isoelectric focusing technique. HRP isoenzymes are combined in
three groups based on their isoelectric points: HRP A (acidic), HRP B and C (neutral and
neutral basic), and HRP D and E (basic) [16,17,18,19].
HRP C is the most abundant and well characterized type of the HRP isoenzymes. HRP C
is a largely α-helical single polypeptide with a ferric heme prosthetic group including two
structural calcium ions and 308 amino acid residues. It is extensively glycosylated with
18-20% carbohydrate content and a weight of 44kDa [16]. The high resolution three-
dimensional structure of HRP C has been elucidated in 1997 using x-ray crystallography
[20]. HRP is an ellipsoid with typical dimensions 6.5nm x 5.4nm x 4.3nm [21]. The pH
optimum of HRP C is in the range of 6.0 to 6.5 and the temperature optimum is in the
ambient range [15].
The general reaction catalyzed by HRP is presented below:
"�#� + �$"� "%&→ �"�# + �$" • [2.6]
Hydrogen peroxide acts as a hydrogen acceptor and AH2 represents a generic hydrogen
donor. HRP specifically catalyzes redox reactions involving hydrogen peroxide as the
oxidant substrate and a relatively large number of possible reducing co-substrates,
including phenols, aminophenols, indophenols, diamines and leuco-dyes and others. The
HRP catalytic mechanism was the subject of extensive study [22,23]. In Figure 2.5, a
general catalytic cycle for peroxidases is depicted. Compound II is more prone to
inactivation and in the presence of excess H2O2 produces a pink inactive compound
(compound III).
— 18 —
Figure 2.5 – The catalytic cycle of HRP C with a generic reducing substrate
Unlike other enzymes, HRP is produced in relatively large quantities because of its many
applications. The enzyme is widely used as a reagent in bioassays (immunoassays,
clinical diagnostic kits), immunohistochemistry, organic biosynthesis, biotransformation,
waste water treatment, biobleaching, targeted cancer therapy and biosensors [24,25,26].
Some relevant characteristics of HRP that makes it an ideal candidate as a model
bioagent in bioactive paper applications are summarized below:
• Readily available, well-known and well-characterized
• Bioanalytical application alone, in bi-enzyme systems and suitable for
conjugation
• Wide range of pH stability and stable at ambient temperature
• Enzyme inhibitor not expected to be present in paper supports
• High catalytic rate
• Wide choice of activity assays: colorimetric, fluorimetric, chemiluminescence,
electrochemical compatible with possible detection methods after immobilization
on paper
Resting State
Compound I
Compound II
H2O2
H2OAH2
AH•
AH• + H2OAH2
Compound III and IV
H2O2
excess
— 19 —
2.2.3 - The solid support: cellulose-based fibrous materials
The solid support where biomolecules are immobilized plays a significant role in the
performance of bioanalytical systems [27]. No standard or ideal support for each type of
immobilization technique has emerged yet [13]. Instead, the comparison of the responses
of different carriers in defined applications has been the traditional approach in selecting
immobilization supports [28,29].
Supports have been classified in terms of their origin (natural or synthetic),
macroconfiguration (fiber, microgranule, microcrystal, capsules, bead, membrane, etc.),
microconfiguration (porous, non porous) or chemical functionality (e.g. bromo-ethyl-
cellulose, diazo-cellulose, etc.). Among the wide range of supports available for
immobilization, porous substrates offer not only a larger surface area for interactions that
allow a larger biomolecule load, but also can potentially provide suitable environments
for biorecognition events [30].
A number of general aspects have to be considered in the selection and design of a
support matrix; they are summarized in Table 2.2. The three main design objectives are
high biomolecule binding capacity, retention of biomolecule functionality and low cost
[30,31].
Cellulose has found a widespread use as support for immobilized biomolecules
[13,32,33,34]. On one hand, the large availability of cellulose in different physical forms,
the hydrophilic character for preserving a suitable microclimate, and the presence of
compatible hydroxyl groups on the surface for derivatization have been considered as
advantageous. On the other hand, susceptibility to microbial degradation, low rigidity,
and nonspecific adsorption have been indicated as disadvantages that rendered cellulose
materials unattractive for some applications [13,32]. Nevertheless, modified celluloses
have been one of the first supports used to immobilize proteins [35].
— 20 —
Table 2.2 - Considerations related to the design of an immobilization support (Adapted
from [36] and [13])
Property Considerations
Physical Strength, compressibility, surface area, shape/form, particle
size, porosity, pore volume, permeability, density, pressure
drop, solubility, rigidity, swelling behavior
Chemical Hydrophilicity, inertness, available functional groups,
regenerability
Stability Storage, biomolecule functionality, mechanical
Resistance Microbial attack, chemicals, pH, temperature, organic
solvents, proteases, contamination
Safety Biocompatibility, toxicity, application-related
Economic Availability and cost, special equipment, reagents, technical
skill, environmental impact, feasibility for scale-up
Reaction Ligand-ligate binding reactions, kinetics, side reactions,
nonspecific interactions, multiplexing, diffusion limitations on
mass transfer, catalytic productivity
Paper has also been used as support in bioanalytical applications. Dry reagent
chemistries [37,38,39] were designed to combine several separation and reaction
functions in one step. Multilayer devices consisting of support, reflective,
analytical, and spreading layers were constructed by entrapment of biological
components. Paper was one of the materials of choice either as a preformed
matrix or as a reflective layer [38,39]. More recently, the use of paper-based
microfluidic patterns as support in bioanalysis has been reported
[40,41,42,43,44]. In the first step of this approach, paper (filter paper or
chromatographic paper) is hydrophobized using a photoresist [40], or printed
polydimethylsiloxane (PDMS) patterns [41], or alkylketene dimer (AKD) [43]
or polystyrene (PS) [44]. Then, defined hydrophilic areas (microfluidic
— 21 —
channels) are patterned using photolithography [40], or plasma treatment [43],
or inkjet printed solvent [44].
2.2.4 - The deposition system: inkjet technology
2.2.4.1 – Technological options for high speed manufacturing of bioactive papers
Mass production of bioactive papers requires close examination of new and
existing technologies in terms of how to bring together functional biomolecules
with cellulosic fibrous supports. Table 2.3 summarizes potential methods for
bioactive paper manufacturing [6,45, 46].
Web forming, coating, and conversion are beyond the scope of this thesis; thus,
these methods will not be examined in detail in this review. On the contrary,
printing of bioactive materials will be thoroughly reviewed in the next few
sections.
Figure 2.6 – Potential technologies for mass production of bioactive papers
Web Forming
Chemical activation
Grafting
Chemo-enzymatic
fiber modification
Immobilization of
active components
Embedding of
active components
Coating and Converting
Surface sizing
Coating
• Spraying
• Hybrid sol-gel coatings
• Dispersion coating
• Curtain
Plasma treatment
Lacquering
Printing
Contact
• Gravure
• Flexography
• Screen printing
• Microspotting
• Softlithography
Non contact
• Photolithography
• Electrospray
• Laser printing
• Continuous flow
• Ink jet printing
— 22 —
2.2.4.2- Bioprinting: printing bioagents on paper
Some conventional and emerging printing techniques have shown potential or
are currently used for depositing solutions of bioactive materials onto solid
substrates [45,47]. Besides the different principles of operation, cost and
performance characteristics distinguish the systems as well. The impact of the
printing process on the ink biological functionality is the most significant factor
in bioprinting. A review and classification of competing bioprinting
technologies will be presented in the next paragraphs.
(a) Contact Dispensing
(a1) Gravure
In gravure printing, ink is transferred from small cells engraved on the surface of
a printing cylinder to the paper (See Figure 2.7). This printing method is
preferred for applications that require high production volumes and high printing
quality, e.g., packaging. No obvious draw backs seemed to be associated with
the gravure printing technique as a method for depositing biomolecules other
than being a contact printing method. Interestingly, no study in the literature has
used gravure printing technology for bioprinting applications. [6].
Figure 2.7 - Gravure
— 23 —
(a2) Flexography
In this technique, ink is transferred from a patterned elastomeric stamp (protrusions
correspond to image areas) to a substrate (See Figure 2.8). Flexography can be used to
print on absorbent and non absorbent materials; hence, it is widely used for printing all
sorts of packaging materials. This method seems to hold promise in bioprinting
applications but has not been explored extensively yet [6].
Figure 2.8 - Flexography
(a3) Screen printing
In this simple technique, the pattern is created by passing ink through openings in a
template applied on the substrate (See Figure 2.9). Screen printing can handle highly
viscous inks and can deliver large amounts of ink to surfaces. It is typically used in fabric
and textile printing. In addition, many examples can be found on applications of this
printing method in biosensor and test strip manufacturing [6,48].
Figure 2.9 - Screen printing
— 24 —
(a4) Microspotting or pin printing
In this technique, introduced in 1996 by Shalon et al. [49], the biological
solution is deposited sequentially by direct contact of a solid, a split pin [50], a
tweezer, or a ring [30] with a solid surface (See Figure 2.10). Microspotting has
been the prevalent system used to produce DNA microarrays [45]. With the
exception of cells [51], its use has not been extended further to other types of
bioactive materials. The minimum volume delivered, which determines the
spatial resolution of the method, is in the nanoliter range, whereas the speed is
limited to 180 spots/pin/min [49]. Limitations of the method include low speed,
high cost and, high risk of contamination. Some specific issues related to split
pins are clogging, tip deformation, and spot uniformity.
Figure 2.10 - Microspotting. (Adapted from [47])
(a5) Microstamping or softlithography
This method, originally developed by Whitesides and coworkers [52,53,54,55]
for printing organic molecules on gold surfaces, is similar to flexography in the
use of an elastomer to transfer the biomolecule to the solid substrate [56]. As
shown in Figure 2.11, the biomolecules are first immobilized on either a
hydrogel stamp (contact processing, CP) or a microstructured elastomer stamp
by reversible adsorption (microcontact printing, µCP) or reversible binding to
capture the molecules (affinity contact printing, αCP). After partially drying, the
— 25 —
biomolecules are transferred by contacting the disposable stamp with the
substrate [11,52,57,58]. Spatial resolutions of less than 100nm have been
achieved [14]. Though simplicity is one of the main advantages of this
technique, it is difficult to control the amount of ink transferred by the stamp
and some authors believe that the lack of humidity during printing may impair
the biomolecule functionality [47]. In addition, the time required for the initial
inking of the stamp can be relatively long [11].
Figure 2.11 - Microcontact printing. (Adapted from [47]).
(b) Non contact dispensing
(b1) Photolitography
In photolithography, a substrate previously coated with a photoactive compound
is illuminated usually by UV light, through a photomask. The UV light creates
reactive sites in the exposed areas of the substrate where the biomolecules in
solution can bind [59]. The general steps involved in the photolithography
process are illustrated in Figure 2.12. The end result is a bioactive layer with the
same shape as a negative of the mask on the substrate. Photolithography allows
printing of all patterns simultaneously with a high spatial resolution (less than
250nm). However, both UV light and the chemical reagents used in the different
steps of this technique can produce biomolecule denaturing [11]. Non-specific
— 26 —
binding can present another issue [59]. In addition, the many steps required by
the photolithographic process limit the number of biomolecules that can be
deposited and the complexity of the printed patterns. Nevertheless, some
researchers have applied photolithography to pattern proteins. For example,
Pritchard et al. [60] immobilized five different antibodies and casein on a silicon
dioxide surface using photobiotin and Pirrung and Huang [61] applied multiple
proteins on BSA derivatized glass support using a photoactive biotin derivative.
Hengsakul and Cass [62] placed biotinylated enzymes on the surface of
polystyrene and nitrocellulose using photobiotin to covalently attach avidin and
Mooney et al. [63] immobilized antibodies on a silicon dioxide surface using
silane to adsorb biotinylated BSA.
Figure 2.12 – Photolithography. (Adapted from [47]).
(b2) Electrospray deposition (ESD)
Two forms of ESD: proximal and distal, have been developed. In proximal ESD
(Figure 2.13(a)), a glass capillary situated close (130-150µm) to a conductive
substrate is exposed to high voltage delivering sequentially spots formed by
micrometric charged droplets of biological solution [47,64]. The technique can
deposit small volumes (in the picoliter range) but the spot size obtained can be
UV
Mask
Biomolecule
Support
— 27 —
relatively large and nonuniform [44]. Distal ESD (Figure 2.13(b)) relies on the
same principle but the glass capillary is situated far (30cm) from the substrate;
hence, the charged microdroplets dry before reaching the substrate through a
dielectric mask. In contrast to proximal ESD, distal ESD creates patterns in
parallel. The technology is considered cost effective for biological materials and
proteins retain their functionality over time due to their dry state [65]. However,
the spot uniformity and density are poor when compared with other bioprinting
methods [45].
Figure 2.13(a) - Distal ESD.
(Adapted from [47])
Figure 2.13(b) - Proximal ESD.
(Adapted from [47])
Morozov and coworkers [66,67,68] introduced and applied distal ESD
techniques for deposition followed by chemical attachment of multiple
biomolecules with a wide range of activities, structures and properties (enzymes,
antibodies, DNA) in a microarray. Only conducting supports can be used in this
printing method and voltage and current need to be carefully optimized to avoid
denaturation. Moreover, for some biomolecules, like certain enzymes,
carbohydrate protectors are required to preserve the biomolecule activity
[66,67].
U
U
— 28 —
(b3) Biological laser printing (BioLPTM)
In biological laser printing [69,70], photothermal and photomechanical effects
induced by focused laser pulses are used to eject discrete spots of a biological
material from a carrier support onto a substrate (See Figure 2.14). The carrier
support contains a UV transparent quartz layer, an absorbing metal oxide
coating layer, and the sample layer (solid, powder, liquid or gel). The receiving
substrate is usually a glass slide.
Figure 2.14 - Biological laser printing. (Adapted from [69])
The technique allows printed spots with diameters as small as 70µm, a spatial
resolution better than 5µm, and a wide deposited volume range from femtoliter
to nanoliter [69,70]. Being an orifice-free design, cross-contamination and
clogging issues are eliminated and it can transfer samples in dry state. It has
been applied for sequentially printing two-dimensional and three-dimensional
patterns of cells and proteins [47,69,70,71]. Even though preliminary results for
bacteria showed no loss of viability after printing, the impact of the UV
exposure and laser thermal energy on the functionality of other types of
biomolecules is still unclear.
Laserquartz
metal oxide
sample
substrate
carrier
— 29 —
(b4) Continuous flow microfluidic printing (CFM)
CFM is a three-dimensional array of microfluidic channels embedded in
polydimethylsiloxane (PDMS) that allows spotting biomolecules in 250µm-side
independently-addressable discrete spots while under continuous flow [72]. The
biomolecule of interest is directly adsorbed on the substrate or captured via
recognition. The system proves advantageous for printing biomolecules that are
not available in high enough concentration or purity. Also, no spot cross
contamination is possible.
Figure 2.15 – Continuous- flow microfluidic printing
(b5) Inkjet printing
(b5.1) Principle of operation, benefits and challenges
Inkjet printing relies on the application of a force to create a high speed liquid
stream (jet) that is ejected through a small orifice (nozzle), resulting in the
formation of uniformly sized and spaced drops [73,74]. A number of actuation
strategies have been devised in inkjet printing [74,75,76]. Continuous inkjet
(CIJ) and drop-on-demand (DOD) systems, either piezo or thermal, are the most
common systems (see Figure 2.16). In CIJ systems, drops are constantly pumped
bioink
microfluidic
channels
printed
spots
— 30 —
through a nozzle; some drops are selectively charged and can be deflected in an
electric field to produce the printed pattern. In contrast, DOD systems only
produce drops when needed, either by exciting a piezoelectric material with a
periodic voltage wave (piezo inkjet) or by superheating the ink until a bubble is
produced that expels a drop from the nozzle (thermal inkjet or bubble jet).
Continuous Inkjet Piezo Inkjet Thermal Inkjet
Figure 2.16 - Main inkjet technologies. (Adapted from [75])
Inkjet printing has been described as a “versatile tool to deposit tiny amounts of
liquid in an extremely defined and controlled manner” [77]. Flexibility, low
cost, easy of mass production, material efficiency, non contact and direct writing
process, speed (up to 25,000 drops/sec/nozzle [74]), good repeatability,
amenable to miniaturization are some of the favorable features associated with
the inkjet technology [71,78,79]. Relatively small dispensed volumes (2pl to 5nl
[74]) and comparatively high spatial resolutions (as high as 2400 d.p.i. (dots per
inch)) are achievable.
In contrast, one of the main challenges in this method seems to be the high
sensitivity to the rheological properties of the ink. Ink viscosity and surface
tension are critical parameters [80,81,82]. Cavitation bubbles [76], undesired
deflected
drops
piezo
sleeve
heater
bubble
— 31 —
satellite drops [76,83], hydraulic cross talk [83], nozzle clogging [75,76,77],
droplet instabilities [78], agglomeration, precipitation or deposition of solutes
[78] are some of the problems that may need to be addressed, depending on the
system.
(b5.2) Piezoelectric jetting parameters: jetting cycle and maintenance operations
In piezoelectric inkjet actuation, a voltage wave is applied to a piezoelectric
element coupled to an ink reservoir to produce a deformation wave that changes
the volume of the reservoir. This volume change causes, in turn, a pressure wave
that, depending on the pressure sign, either ejects a drop from the ink reservoir
trough a nozzle or refills the reservoir with ink. Figure 2.17 is a simplified
scheme of the cross section of a single piezoelectric drop-on-demand ejector
with a shear-mode flat-drive plate actuator design [84,85,86,87]. The
piezoelectric jetting cycle can be described in terms of the different phases of
the driving waveform. Figure 2.18 illustrates a generic jetting cycle.
Figure 2.17- Cross section of a single piezoelectric drop-on-demand ejector
Lee [84] defined five adjustable operating parameters for reliable jetting in DOD
devices: drive pulse amplitude (related to the nozzle voltage), drive pulse shape
(related to waveform shape and pulse width), internal pressure level (related to
the meniscus vacuum), drop ejection rate (related to the firing frequency), and
fluid fill level. In addition, printhead priming and cleaning are some
maintenance operations executed before, during, and after printing that can
Piezoelectric element
NozzleInk channel
Connection
with
ink reservoir
— 32 —
affect the jetting performance. The type, frequency, and duration of these
maintenance operations should be adjusted to improve jetting reliability.
a) Equilibrium
b) Drawing ink from reservoir
c) Ejection and drop formation
d) Drop break off and jet withdrawal
Figure 2.18- Generic piezoelectric jetting cycle
Vo
lta
ge
Time
Standby
Pulse width
Amplitude
Vo
lta
ge
Time
Ink
filling
Vo
lta
ge
Time
Drop
ejection
Vo
lta
ge
Time
Recovery
— 33 —
(b5.3) - Jetting performance criteria: feasibility and reliability
Jetting feasibility addresses the question: can drops be jetted? Jetting reliability
addresses the question: can repeatable drops be constantly jetted? Therefore, a
reliably jettable bio-ink is a biologically active ink with adjusted rheological
properties for inkjet printing that can produce equally spaced and equally sized
drops on demand (either in intermittent or continuous operation) with minimum
maintenance operations until all the ink in the reservoir (cartridge) is consumed.
Kang [82] defined four performance criteria to evaluate jettability: ink-material
compatibility, drop formation, orifice clogging, and faceplate wetting.
Ink-material compatibility: The materials in physical contact with the ink
during the printing process (cartridge, printhead) as well as during idle periods
(cleaning pad) should be inert with respect to the ink. Chemical reactivity with
the printead materials (corrosive inks, organic solvents) [84] or materials that
can alter the activity of bio-inks (inhibitors) are examples of possible
incompatibilities.
Drop formation: Stable jetting is achieved when monodisperse drops, with the
same speed, size, direction of travel, and free of satellites are formed [84]. The
fundamental dynamics of the drop formation process is poorly understood. Thus,
finding a reliable printing window relies on a trial and error process [84,88]. For
a given ink and printhead, drop formation can only be modified through changes
in the parameters of the jetting cycle explained in section (b5.2).
Nozzle clogging: Inkjet inks should be clean and free of particles, sediments or
aggregates that can eventually obstruct the ejection hole. Nozzle clogging
problems can be mostly eliminated by filtering the ink before filling the
cartridge; however, during printing, dry ink deposits can form and block the
jetting orifice causing unstable, misdirected, or no drop ejection. The problem
tends to appear randomly at any nozzle and during no particular stage of the
— 34 —
printing process. Suitable cleaning cycles and an addition of humectants to the
bio-ink formulation can help minimize the issue.
Faceplate wetting: The ink layer wetting the external surface surrounding the
nozzles is a key factor in jettability studies. Large amounts of ink on the
faceplate demand a larger driving amplitude, can impair drop ejection and
produce flooding or dripping. In addition to the design of printheads with highly
hydrophobic faceplate surfaces, the problem can be controlled by application of
an internal negative pressure to the ink circuit (meniscus pressure) [84].
(b5.4) Non conventional applications of inkjet technology
Initially developed for office printing, inkjet printing is a promising and cost-
effective technology to incorporate functionalities into materials [45,74,75,89].
Table 2.3 summarizes current and emerging areas of application of inkjet
technology not related to the graphic arts. Sirringhaus and Shimoda [84]
highlighted four areas of challenge in the application of inkjet technology to
deliver functional materials:
• Ink formulation: evaluation of new additives for control of rheological
properties, compatibility with the printer ink circuit, and preservation of
the functionality of the main active component
• Print head and print system design: compatible with new inks, reliable
and stable jetting, no clogging, repeatable drop volume, and highly
accurate droplet positioning
• Substrate choice and preparation: control of ink spreading and surface
chemistry
• Control of solvent evaporation: control of ink drying at the nozzles
(jettability) and on the substrate (position, structure and profile)
— 35 —
Table 2.3 - Non conventional applications of the inkjet technology
Application Ink References
Microdispensing
• Biomolecules
• Polymers
DNA
Cells
Microorganisms
Antibodies
Hormones
Enzymes
Protein
Polymers
[74,90,91,92,93,94,95,96]
[84,97,98,99]
[100,101]
[102, 103,104]
[105]
[106,107, 108,109, 110,111,112,113, 114]
[77,80, 115]
[80,106]
Chemical analysis
• Drug discovery [50]
• Biosensors
• Combinatorial chemistry
• Mass spectrometry
[102, 104,105,116]
[117]
[118,119,120]
Chemical synthesis
• Nucleic acids
• Peptides
• Polymers
Base
Amino acids
Monomers
[96]
[121]
[122]
3-D manufacturing
• Tissue engineering
• Structural parts
Cells
Ceramics
[93,123,124]
[76]
Microassembly
• Organic electronic
components
• Light emitting diodes
• Liquid crystal displays
• Soldering
• Circuit boards
[83]
[125]
[77]
[126]
[127]
— 36 —
(b5.5) Enzyme inkjet printing
Some researchers have successfully printed enzymes using inkjet technology.
Table 2.4 details the printing systems, biocatalysts, and supports used by these
researchers and their intended applications.
Table 2.4 - Previous work on inkjet printing of enzymes
Ref. Inkjet Enzyme Support Application
[106] piezoelectric GOD, urease
ISFET sapphire biosensor
[107] electrostatic GOD carbon electrode biosensor
[108] thermal HRP papers , plastics bioanalysis
[111] thermal GAL polyester sheet exploratory
[110] thermal GOD glass coated with PEDOT-PSS polymer
biosensor
[112] thermal HRP glass coated with PEDOT-PSS polymer
biosensor
[109] piezoelectric flow- through
GOD Au sputtered glass slide biosensor
[113] piezoelectric flow -through
GOD
LOD
SAM modified Au coated surface
multisensor
[44] piezoelectric GOD
HRP
paper multisensor
Note: GOD, LOD, HRP, GAL, ISFET, PEDOT/PSS and SAM stand for glucose oxidase, lactate oxidase,
horseradish peroxidase, galactosidase, ion selective field effect transistor, poly (3,4-
ethylenedioxythiophene/polystyrene sulfonic acid, and self assembled monolayer, respectively. Drops in
piezoelectric flow through systems are generated perpendicular to a flow passing through a cell as opposed
to drops generated from a limited volume of ink contained in a cartridge.
Few studies have attempted to pattern enzymes on cellulosic fibrous supports. In
a pioneering research by Roda et al. [108], a bio-ink containing HRP, buffer,
and surfactant was deposited onto various solid supports using a commercial
thermal inkjet printer. The solid supports included cellulose papers with basis
— 37 —
weight ranging between 30 and 80g/m², cellulose filter paper, nylon sheet,
photographic gelatin paper, tissue paper, and inkjet transparency film. The
authors reported that the best intensity and spatial distribution in the
chemiluminescent response was obtained with the permeable paper supports.
Moreover, the fast diffusion of chemiluminescent substrate in the low basis
weight paper resulted in a faster signal development. The other non permeable
supports produced detection problems due to enzyme washout.
More recently, Abe et al. [44] used a piezoelectric inkjet printer to deliver a
solution of GOD and HRP to one of the sensing areas of a microfluidic
multianalyte chemical sensing paper for simultaneous quantitative colorimetric
detection of pH, protein, and glucose in urine. The authors observed some
problems of inhomogeneous color distribution, specifically in the sensing area
where GOD and HRP were printed, that will require a significant improvement.
2.2.5 - Paper-biomolecule attachment: passive adsorption
2.2.5.1 – Immobilization strategies
Immobilization is defined as the localization of biomolecules in a microspace with a
retained functionality [128,129]. Despite the added cost, immobilizing biomolecules onto
supports proves to be beneficial. Immobilization allows recovery and reuse of the
bioagent in some biocatalysis [32, 130 ] and provides a high local concentration of
bioagent in microarrays [ 131 ]. It offers a controlled microenvironment that can
selectively alter properties of the biomolecule [27] and may result in improved storage
and operational stability [13,27,131]. However, challenges such as loss of the
biomolecule activity and mass transfer limitations may have to be solved for particular
combinations of support-immobilization-bioagent system.
The choice of immobilization strategy depends on the type of application, support, and
biomolecule. Table 2.5 lists different objectives for immobilization in biotransformations
and bioanalysis.
— 38 —
Table 2.5 – Immobilization objectives [130,132]
Biotransformations Bioanalysis
• Enhanced storage and operational
stability
• Reuse and recovery of biomolecule
• Reduce biocatalyst cost
• Retention of activity
• The carrier “dilutes” the activity
• High biomolecule surface density
• High sensitivity
• Non-specific interactions absent
• Full retention of protein
conformation and activity
• The carrier “concentrates” the
activity
The main immobilization strategies have been classified as chemical, physical and
biological [13,132]. Chemical methods rely upon the formation of one or more covalent
bonds between the bioagent and the support. Conversely, physical methods involve the
confinement of biomolecules by means of physical forces (like electrostatic, ionic or
hydrophobic forces) without formation of covalent bonds. In biological methods, the high
biochemical affinity between binding pairs is used to attach biomolecules to surfaces.
Figure 2.17 schematically depicts the different immobilization strategies classified
according to the nature of the attachment between the biomolecule and the support
[13,132,133].
Some favorable aspects of physical immobilization methods are:
• Less risk of damage to the biomolecule structure [134a]
• Less expensive since few reagents are needed
• Simpler due to less experimental steps
• No chemical modifications are introduced to the support or biomolecule
• Reversible in most of the cases and amenable to regeneration
— 39 —
Physical Methods
Adsorption Entrapment Confinement
Chemical Methods Biological Methods
Cross-linking Covalent bonding Bioaffinity
Figure 2.19 - Immobilization strategies. (Some schemes were adapted from [133])
Some of the disadvantages of physical immobilization methods include:
• Attachments are generally weak; hence, not very stable or permanent
• Progressive loss of biological activity
• Desorption and leakage of biomolecules from the support
• Increased nonspecific binding
• Increased cost caused by the overloading of the support with biomolecule
(compensation for losses by inactivation and leakage) [135]
• Steric hindrance by the support
The particular advantages of chemical immobilization strategies are:
• Biomolecule lifetime is greatly improved
• Permanent and stable attachment of the biomolecules to the support
— 40 —
• Minimal leakage of the biomolecules during use
The disadvantages of chemical immobilization strategies are:
• The coupling reaction can compromise the activity of the biomolecule
• The method is more expensive and complex
• Activated surfaces can denature proteins if too closely bound to the surface
• Irreversible and the biomolecule cannot be recovered
(a) Adsorption
Immobilization by adsorption involves the direct binding of the biomolecule to a surface
by non-covalent bonds. The strength of these bonds varies from weak van der Waals
interactions, stronger hydrophobic effects, up to very strong ionic bonds. For non polar
surfaces, adsorption is driven by hydrophobic effects. This strong physisorption can lead
to progressive changes in the biomolecule (generally a protein) tertiary structure and may
result in denaturing. For hydrophilic surfaces, adsorption is driven by van der Waals
forces or ionic and hydrogen bonding interactions. In this type of adsorption, factors such
as pH, ionic strength, and the presence of surfactants will influence the protein binding.
(b) Entrapment
This immobilization method relies on the frequently large difference in dimensions found
between binding partners. Gel entrapment and microencapsulation are examples of this
approach. In the former, the biomolecule is enclosed within a semipermeable polymer
membrane in the form of microcapsules with 1 to 100µm in size [136]. In the latter, the
biomolecule is mixed with a monomer solution, which is then polymerized to form a
highly cross-linked water insoluble gel. The result is a three dimensional porous network
that physically entraps the biological component. In both cases, the porosity of either the
microcapsule wall or the gel lattice is controlled to ensure that the structure is tight
enough to prevent leakage of the biomolecules, but sufficiently open to allow free
passage of small analytes.
— 41 —
The distinction between binding (physical, chemical or biological) and entrapment could
be vague. Sheldon [130] considers binding as the immobilization of biomolecules on
prefabricated supports either in internal or external surfaces, whereas the simultaneous
synthesis of a support network in the presence of the biomolecule is considered
immobilization by entrapment.
The advantage of entrapment over adsorption is the reduced washout of the biomolecules,
although sometimes leakage of the biomolecules can occur due to the broad distribution
of pore sizes observed in some gels. Unfortunately, the support acts as a barrier to the
mass transfer of the binding partners. Due to diffusion limitations, only small analytes
can be detected with a reasonable response time. Another problem that may contribute to
the low diffusion rates is the progressive clogging of the pores with other sample
components.
(c) Confinement
The biomolecules are deposited on a semipermeable membrane with controlled porosity
that allows the free transport of small analytes or products inside and outside the support
but impedes the movement of the enzyme [135]. Two of the most commonly used
supports are dialysis and ultrafiltration membranes.
(d) Cross-linking
Bifunctional or multifunctional reagents are used to create intermolecular covalent bonds
between the biomolecules and/or to the support. Dialdehydes, such as glutaraldehyde, are
widely used as cross-linkers for protein immobilization. The severe conditions required
for some cross-linking reactions can lead to a significant loss of protein functionality and
poor reproducibility [130]. Cross-linked enzyme crystals and cross-linked enzyme
aggregates are examples of recent applications of this immobilization strategy to
biotransformations.
— 42 —
(e) Covalent binding
This method of immobilization involves the formation of a covalent bond between the
biomolecule and the solid support. The bond is normally formed between functional
groups present on the surface of the support and functional groups belonging to the
biomolecule. Thus, first the solid support is activated and then the coupling of the
biomolecule to the surface takes place.
A large number of reactions are available. Amino, carboxyl, hydroxyl, disulphide, and
sulfhydryl functional groups present in the biomolecules (nucleic acids, enzymes,
antibodies, etc) are suitable targets for covalent bond formation with carboxyl, hydroxyl,
amino, and sinalol groups present on the support surface.
The introduction of a recognition element to the support is an effective way to avoid
leakage and non-specific binding. However, it is important to choose a method that will
form covalent bonds with non-essential sites in the biomaterial, avoiding the blockage of
the active sites involved in the biorecognition event. Also, in order to minimize protein
denaturing/inactivation, the reaction leading to the formation of chemical bonds has to be
performed under mild conditions, i.e., low temperatures, low ionic strengths and pH
levels in the physiological range.
(f) Bioaffinity
In this immobilization strategy, the strong non-covalent bond (very high affinity
constants, see Table 2.1 before) between the binding partners is exploited to attach the
biomolecules to the surfaces that have the complementary affinity moiety. The avidin-
biotin system, the His-Tag system, DNA-directed immobilization, and affinity capture
ligand system are extensively used in bioaffinity applications [132]. This method has the
distinctive advantage of keeping biomolecules in the correct orientation and produces
homogeneous, highly specific, and reversible attachments. However, it involves
additional steps: one of the binding partners has to be immobilized on the support and the
remaining binding partner be conjugated or expressed in the biomolecule, adding more
complexity and cost.
— 43 —
2.2.5.2 - Paper-enzyme interactions during adsorption
Understanding and tailoring the interactions between the substrate surface and
the biomolecules (paper and enzymes in particular) is critically important in
developing bioactive papers [137]. Protein-surface interactions govern the
distribution, binding behavior, biological activity, and stability of the
immobilized biomolecules, significantly affecting, in turn, the functional
performance of the bioactive paper.
Leckband and Israelachvili [138] wrote a comprehensive literature review about
the different intermolecular forces acting in biological systems. The review
emphasized the higher complexity of biological interactions involving
complementarity, non equilibrium, and non linearity. Due to their amphiphilic
character, proteins interact with surfaces mainly through hydrophobic and ionic
interactions. In most of the cases, higher adsorbed amounts of protein are
observed on hydrophobic surfaces [139,140,141,142,143]. Upon adsorption, the
interactions between the sorbent and the protein can produce changes in the
structure and conformation of the bound enzymes that can result in different
levels of inactivation [144,145,146]. It has been suggested that stronger
interfacial interactions (commonly hydrophobic) can lead to extensive protein
unfolding.
Although techniques for characterizing adsorption behaviour of biomolecules on
surfaces are well developed, scarce information is available in the literature
about interactions of biomolecules, in particular enzymes, and paper surfaces for
biosensor applications.
2.2.6-Detection system: colorimetric
After the biorecognition event is triggered, a detection system is needed to
interrogate the biomolecular interaction. Different detection systems can be
envisioned in bioactive papers:
— 44 —
Figure 2.20 – Detection systems
The reporting function is generally achieved by incorporating labels or
indicators in the system that change color (chromophores [147], quantum dots
[148], gold nanoparticles [149]), fluorescence (fluorophores, molecular beacons,
fluorescent proteins), or luminescence (luminophores), or radioactivity
(radiolabels) upon binding of the analyte to the bioagent. Chromophores in
particular, have the advantage of giving a visible indication that does not require
expensive or complex instrumentation for detection [150]. However, labelling
imposes additional cost and time and can produce interferences. Relatively few
labels are available in comparison to the number of bioagents and reactions to be
detected [151].
Quantitative detections, on the other hand, typically require a transduction
element. The main purpose of the transduction element is to convert an observed
change (physical or chemical) into a measurable signal with a magnitude
proportional to the concentration of the analyte [134b]. In many cases, direct
quantitative detection (label-free) is achieved by measuring changes in electrical
(potentiometric, amperometric or conductometric), mass (piezoelectric, acoustic
wave), heat (calorimetric), or optical (luminescent, fluorescent, reflective,
interferometric, ellipsometric, surface plasmon resonance and waveguide)
properties [133,152,153,154].
Qualitative
passive detection
REPORT
Quantitative
passive detection
REPORT MEASURE
Reactive
detectionREPORT MEASURE
TAKE
ACTION
— 45 —
VTT’s literature review [6] estimated that electrical and optical transducers will
be the preferred transduction principles in bioactive papers. In particular, the use
of fluorophores in bioactive papers will be challenging because of the complex
sample handling issues and potential interference from the autofluorescence of
the cellulosic fibers. Ideally, fully portable bioactive papers are envisioned as
self-contained analytical devices with the detection system embedded in the
paper (instrument-less readout) or, alternatively, make use of portable detectors.
In applications where expert evaluation of the results is required (e.g. clinical
diagnosis), the ability to digitize and transmit readouts offsite to the specialist
will be an added post-detection benefit [42].
2.3 - EVALUATING BIOACTIVE PAPER PERFORMANCE
After completing the design of a bioactive paper, its functional performance has
to be assessed. Some of the performance indicators used to evaluate
bioanalytical applications and their definitions are [134c,155]:
• Selectivity/Specificity [134c]: Given mainly by the bioagent and its ability to
discriminate the analyte (or group of analytes) of interest from other species
present in the sample.
• Range of analysis [134c]: the analyte concentrations covered by the calibration
curve.
• Linear range [155]: “Concentration range over which the intensity of the signal
obtained is directly proportional to the concentration of the species producing the
signal”.
• Lower limit of detection [155]: “the analyte concentration derived from the
smallest measurement, LLD, that can be detected with reasonable certainty. The
value of LLD is given by the equation:
LLD LLD LLD LLD = = = = XXXXBiBiBiBi++++++++++++++++ ±±±± k k k k σσσσBiBiBiBi [2.6]
Where,
XXXX+ BiBiBiBi mean of the blank measures
— 46 —
σσσσBiBiBiBi standard deviation of blank measures
kkkk numerical factor chosen according to the confidence level desired”
• Sensitivity [155,156]: “The slope of the calibration curve in the linear range. If
the curve is in fact a 'curve', rather than a straight line, then of course sensitivity
will be a function of analyte concentration or amount”.
• Response Time: time elapsed between moment the paper contacts the sample and
the moment readout is taken. This measurement is affected by factors such as the
sample preparation time, the simplicity of the measurement and the time needed
to reach equilibrium or constant measure [134c].
• Lifetime: time span the bioactive paper can be stored after manufacturing before a
deterioration of the performance is observed. Also known as stability.
• Precision [155]: “The closeness of agreement between independent test results
obtained by applying the experimental procedure under stipulated conditions. A
measure of precision (or imprecision) is the standard deviation”. In the context of
bioactive papers, precision will be affected by batch-to-batch variability in the
bioagent, variability in the paper support (anisotropy, manufacturing conditions)
and variability in the test conditions (pH, temperature and humidity). A suitable
replication strategy should be used to minimize random errors.
• Accuracy [155]: “The closeness of the agreement between the result of a
measurement and the true value”. Suitable standards and controls should be used.
• Repeatability [155]: “The closeness of agreement between independent results
obtained with the same method on identical test material, under the same
conditions”
• Reproducibility [155]: “The closeness of agreement between independent results
obtained with the same method on identical test material but under different
conditions”
• Signal to noise ratio: Ratio between mean measurement and the deviations
affecting the measurement.
• Background: The analytical response of the bioactive paper in the absence of
some of the elements of the biorecognition (analyte, binding partner, indicator).
— 47 —
Non-specific binding, i.e., the binding of the bioagent to non-target binding sites,
is one of the most important sources of background in bioanalysis.
• Interferences [152]: “A systematic error in the measure of a signal caused by the
presence of concomitants in a sample”. In the context of bioactive papers the
concomitants are substances in the bioagent source, in the paper composition, in
the deposition system, in the sample matrix, or in the measurement environment
that affect the output of the measurement.
Performance factors of bioactive papers will likely be affected by the type and size of
analyte [157], sample matrix [158a], type and amount of bioagent [158b], spot size and
density [159], support [160], immobilization method [132], label, and/or transducer type
[132].
2.4 - CONCLUSIONS
The field of bioactive papers is still in its infancy. Many aspects of the design of
bioactive papers are yet to be explored. The use of paper and other cellulosic materials in
bioanalytical systems is attractive not only because a bioanalytical system incorporating a
low-price readily-available commodity is easier to scale up and commercialize, but also
because of the potential benefits of higher value and product differentiation for the pulp
and paper industry. The optimum combination of components (bioagent-deposition-
immobilization-support-detection) for a particular bioanalytical application has been
determined mainly empirically. Systematic studies of the impact of support properties,
immobilization strategy, and deposition parameters on the performance of bioanalytical
systems are few in the literature.
The use of cellulosic fibers and paper in conjunction with biological molecules is rare.
Even though modified cellulose substrates for covalent binding of bioagents have
received considerable attention in the past, scarce research has been done recently
containing paper or cellulose supports involving simple physical immobilization
techniques. Moreover, few attempts to apply multi-bioagents on cellulosic supports have
been reported, even fewer involving paper as a substrate.
— 48 —
Proteins are among the most difficult bioagents to print due to their fragile nature and
high susceptibility to denaturation. Enzymes in particular appear to be one of the most
challenging bioagents because printing or immobilization can lead to inactivation.
Nevertheless, many enzymes such as some oxidoreductases are well known and well
characterized in solution state. That makes them ideal subjects for investigation in solid
phase systems. Despite the considerable effort needed to optimize bioanalytical systems
involving enzymes, it is likely that providing a scientific roadmap for the development of
a bioactive paper containing a model enzyme bioagent would be an useful guide for
extension to other enzymes and less sensitive bioagent systems.
Among the deposition methods reviewed, the absence of physical contact with the
support is the most valuable feature in non contact bioprinting. Within this particular
group of technologies, inkjetting can accurately and rapidly place small volumes with the
added advantage of posing low demands on the support properties. Nonetheless, ink
formulation in this printing system is crucial. Most research on inkjet printing of enzymes
is exploratory and focused on printing feasibility. The specific challenges associated with
the formulation of enzyme-containing inkjet bio-inks have not been explicitly studied.
Clearly, research focused on inkjet printing reliability is needed to advance towards high
speed manufacturing of bioactive papers.
Physical adsorption is the simplest immobilization strategy for attaching biomolecules to
surfaces and has been extensively investigated over the years. However, in many protein-
paper systems the interactions between proteins, enzymes in particular, and paper
surfaces are poorly understood. A characterization of the adsorption behaviour of
enzymes on cellulosic fibrous supports should be undertaken to identify attributes of the
paper support that determine bioagent attachment. These key paper attributes will allow
to better engineer the paper surface to hold and maintain bioactivity.
The chapters that follow this literature review will address some of the
unexplored areas in bioactive paper research through the study of a model
system involving the solid phase bioanalysis of H2O2 using the enzyme
horseradish peroxidase.
— 49 —
Chapter 3
EXPERIMENTAL APPROACH
The detailed information about the materials, experimental procedures, and
measurement techniques used in this thesis study are summarized in this chapter.
3.1-BIO-INK FORMULATION
3.1.1-Enzyme
Horseradish peroxidase, HRP C, (EC 1.11.1.7, type VI, from amoracia rusticana,
highly stabilized, essentially salt free, 200-300 U/mg, Rz~3.0) was purchased
from Sigma (Oakville, ON, Canada) and used without further purification.
3.1.2-Chromogenic enzyme substrate
ABTS, 2, 2'-azino-di-(3- ethylbenzthiazoline) 6-sulphonate, was purchased from
Sigma and dissolved in 100mM potassium phosphate buffer (pH 5).
3.1.3-Buffer
The enzyme solutions were in all cases prepared in a 40mM potassium
phosphate (Sigma-Aldrich) buffer (pH 6.8) solution.
3.1.4-Additives
Glycerol (Merck anhydrous pure); EG, ethylene glycol (Caledon); PVA,
polyvinyl alcohol (Celvol 107); CMC, sodium carboxymethyl cellulose (Procell
CMC 813LZR); and three PEGs, polyethylene glycols with molecular weights
of 200Da, 2000Da and 20000Da (Fluka BioChemika Ultra) were tested as
viscosity modifiers. A non ionic surfactant, Triton X-100 (Sigma-Aldrich), was
incorporated as a surfactant.
— 50 —
3.1.5-Liquid vehicles formulations
Different liquid vehicles for the enzyme-chromogen system were prepared
according to the formulations detailed in Table 3.1.
Table 3.1- Liquid vehicles formulations
Formulation Code
Viscosity Modifier
Range of doses
Surfactant Buffer Humectant
Control 1 None None
40mM KH2PO4 pH 6.8
None
Control 2
0.1wt.-% Triton X-
100
1 PEG 200 Da 0-50wt.-%
2 PEG 2,000 Da 0-30wt.-%
3 PEG 20,000 Da
0-30wt.-%
4 EG 0-80wt.-%
5 Glycerol 0-70wt.-%
6 PVA 0-6wt.-%
7 CMC 0-1wt.-%
8 CMC 0.5wt.-% 10wt.-% Glycerol
3.1.6-Standard bio-ink formulation
A 50U/ml (pyrogallol units) HRP enzyme and 18.2mM ABTS solution was
prepared in a 40mM potassium phosphate buffer (pH 6.8) containing 0.1 wt-%
Triton X-100 as a surfactant, 10 wt-% glycerol as a humectant and a 0.5 wt-%
carboxymethyl cellulose as a viscosity modifier.
— 51 —
3.1.7-Viscosity measurements
Dynamic viscosity was measured in triplicate at room temperature (23°C) using
a capillary viscometer # 200 calibrated against deionized water.
3.1.8-Surface tension measurements
The surface tension of the different liquid vehicles was measured with a
tensiometer Sigma 700 (KSV Instrument Ltd.) based on the Wilhelmy plate
principle [161]. The selected measurement conditions were:
• speed up/down 20 mm/min
• wetting depth 6 mm
• integration time 300 sec
• measuring time 30 min
• temperature 23°C
• repetitions 6
• solid glass
3.2-SOLID SUPPORTS
3.2.1- Fibrous supports for printing
3.2.1.1 - Commercial papers
In the next page, Table 3.2 summarizes the different commercial papers used as solid supports in
this thesis.
— 52 —
Table 3.2. Commercial paper supports
Code Type General use
A Standard cellulose chromatographic paper chromatography
B Uncoated, mixture of thermo-mechanical, groundwood, kraft and recycled fibers
light weight coating, LWC, base paper
C Uncoated, 30% post-consumer fibers office, printer, copiers
D Uncoated, wood-free office, laser printer
E Color copy cover premium color copies
F Coated grade with calcium carbonate pigment and latex binder
premium offset prints
G Coated inkjet grade with silica pigment and polyvinyl alcohol binder
premium inkjet prints
H Cast coated photographic quality prints
3.2.1.2 –Handsheets with increasing hydrophobicity
Fibers either beaten or both beaten and internally sized (see details of the fiber
treatments in section 3.2.2.2.2) were formed into handsheets with 60g/m² basis
weight in tap water following the standard test method Tappi T220. The
handsheets were air-dried in a conditioned room (23 ± 1°C temperature and 50 ±
2% relative humidity, according to the standard Tappi T402) and then oven-
dried for 5 minutes at 105°C. The handsheets were calendered in the lab in a
Beloit-Wheeler (Beloit-Wheeler, USA) calender at 80°C and under 62kN/m
linear pressure. The basis weight, thickness, and degree of sizing (Hercules
Sizing Test, HST, 80% reflectance, 1% formic acid) of the handsheets were
tested according to the standard methods Tappi T410, T411 and T530,
respectively. In addition, the surface energy of the handsheets was characterized
using the method described in section 3.2.2.3.2.
— 53 —
3.2.1.3 - Coating layer
A thick layer of coating formulated with 10pph of (unit conventionally used in
paper coating formulations, the reference is 100 parts of pigment) a styrene-
butadiene (SB) latex (Styronal BN4606, BASF) and 100pph of a ground calcium
carbonate (Hydrocarb 90, Omya) pigment was applied on one side of a copier
grade transparency using a rod #16 in a laboratory coater, Endupap Universal
Coating Machine developed by CTP (Centre Technique de l’Industrie des
Papiers, Cartons et Celluloses de Grenoble), operating at a speed of 3 m/min and
at a drying temperature of 200°C for 1 minute. Subsequently, the coated side
was inkjet printed with the enzyme bio-ink containing Amplex Red (See details
in section 3.6.2).
3.2.2 -Supports for adsorption
3.2.2.1 – Model sorbents
Microgranular cellulose (C6413, Sigma), ground calcium carbonate (Hydrocarb 90,
Omya), rayon filament yarn (RT700, Cordenka), styrene-butadiene, SB, latex (Styronal
BN4606, BASF), and polystyrene beads (390nm average diameter, Bangs Laboratories)
were used as model supports for HRP adsorption experiments. The rayon filament yarn
was ground using a Thomas® Wiley® Mini-Mill to reduce the size of the fibers to
0.26mm arithmetic average length or 0.35mm length weighted average length (measured
using FQA Fiber Quality Analyzer). Liquid SB latex was deposited on Parafilm using a
pipette, spread with a glass rod and air-dried to form a solid film. Then the latex film was
cut into small pieces (ca. 5mm size) with a scissor. The polystyrene beads were dialyzed
for 48h against 40mM KH2PO4 buffer at pH 6.8 using a 20kDa molecular-weight cut off
(MWCO) Slide-A-Lyzer Dialysis Cassette (Pierce) made out of a low-binding
regenerated cellulose membrane. The dialysis process helped remove additives
(surfactants and biocides) that can denature the enzyme. The sorbents were characterized
in terms of surface energy (details in section 3.2.2.3.1), zeta potential (details in section
3.2.2.3.3), topography (details in section 3.2.2.3.4), and specific surface area (details in
section 3.2.2.3.5).
— 54 —
3.2.2.2- Treated fibers
3.2.2.2.1- Beating
Dry commercial BKSW, bleached kraft softwood pulp fiber (Tembec), was
soaked overnight in distilled water and was beaten to 450ml Canadian Standard
Freeness with a PFI mill to increase the specific surface area.
3.2.2.2.2 -Internal sizing
A 1 wt-% dispersion of beaten fibers in water was sized (synonym of
hydrophobized used in papermaking) using an internal sizing agent of either (A)
a rosin-based size (mixture of resin acids mainly composed of abietic acid) or
(B) an alkyl ketene dimer, AKD, reactive size. In case (A), 0.8 wt-% and 1.6 wt-
% doses of 1 wt-% cationic dispersed rosin size (Ultra-pHase® 35, Hercules)
solution was first added to the fiber dispersion in distilled water, followed, after
30s, by aluminum sulfate. During the sizing process, the pH and the rosin to
aluminium sulfate ratio were kept constant at 4.5 and 1, respectively and the
dispersion was continuously stirred for 8min before filtering and washing with
tap water to obtain neutral pH fibers. In case (B), a 0.3 wt-% dose of AKD
(Hercon® 115, Hercules) was added to a 1 wt-% dispersion of beaten fibers in
tap water. The dispersion was continuously stirred for 8min and kept at neutral
pH before filtering and washing with tap water. The sized fibers were divided
into two parts: one part was formed into handsheets (see section 3.2.1.2), and the
rest was oven-dried for 5 min at 105°C and were directly used in adsorption
experiments. Treated and untreated fibers were characterized in terms of surface
energy (see details in section 3.2.2.3.1), zeta potential (see details in section
3.2.2.3.3), topography (see details in section 3.2.2.3.4), and specific surface area
(see details in section 3.2.2.3.5).
3.2.2.2.3- TEMPO-mediated oxidation of ground rayon and BKSW fiber
The oxidation of cellulosic fibers and rayon was conducted at room temperature
following the method detailed by Kitaoka et al. [162]. The sample (10 g dry
— 55 —
basis) was suspended in a solution containing 0.025g of TEMPO, 2,2,6,6-
tetramethyl-1-piperidininyloxy radical (purified by sublimation, 99%, Aldrich),
0.25g of sodium bromide (≥99%, SigmaUltra), and 750ml of deionized water.
The pH of the suspension was adjusted to 10.5 by adding 0.5M NaOH solution.
The reaction began with the slow addition of 2.42 milimoles of 10.5% sodium
hypochlorite solution (reagent grade, available chlorine >4%, Sigma-Aldrich).
The reaction proceeded for about 2 h, during which the pH was kept at 10.5 by
small additions of 0.5M NaOH solution. When the pH remained practically
unchanged, the reaction was stopped by adding a few ml of ethanol to the
suspension. Finally, the oxidized fiber was filtered and washed with deionized
water until the filtrate was free from alkali.
3.2.2.2.4- Surface charge modification using polyelectrolyte
90mg of poly-(allylaminehydrochloride), PAH, (Aldrich) and 0.585g of sodium
chloride (Caledon) were added to a liter of distilled water and the solution pH
was adjusted to the value of 6.5 using a phosphate buffer of pH 8 . Ground rayon
yarn equivalent to 3g oven dry weight was incorporated into the previous
solution while stirring and was left in contact for 20min. After filtering through
a 25µm-aperture nylon mesh, the rayon fibers were washed three times with
1000ml of 0.01M sodium chloride solution.
3.2.2.3 – Characterization of model sorbents and treated fibers
3.2.2.3.1 – Surface energy of model sorbents
The total surface free energy (γT) of the model sorbents, as well as the dispersive
(γD), polar (γP), electron-acceptor (γ+), and electron-donor (γ-) components of
the surface energy, were obtained from the literature. Table 3.3 summarizes the
information.
— 56 —
Table 3.3 - Surface energy of model sorbents as reported in the literature
Solid Method γ
T γD γ
P γ+ γ
- Ref.
mN/m mN/m mN/m mN/m mN/m
PS Contact angle 42 42 0.5 0 1.1 [163]
PS GvOC theory 38.2 38.4 0.026 0.219 [164]
PS 40.57 38.39 2.17 [165]
PS Harm. mean 42.6 38.4 4.2 [166]
PS Geom. mean 42.0 41.4 0.6 [166]
PS Molt. polymer 40.7 33.9 6.8 [166]
PS DCA 39.1 34.8 4.3 [167]
MCC Contact angle 53.9 36.6 17.3 [168]
MCC IGC 4 0 [169]
MCC IGC 40.3 [170]
MCC Wicking 57.18 52.94 4.24 0.11 41.70 [171]
AC Wicking 58.98 54.49 4.49 0.11 47.83 [171]
AC IGC 35.6 to 49 42 [172]
AC IGC 45.1 [173]
AC* DCA 72.8 22.44 50.38 [174]
AC* DCA 50.06 23.75 26.31 [174]
AC Contact angle 67 33 34 [175]
GCC IGC 68.6 [170]
kaolin DCA 64.3 32.4 31.9 [176]
PCC DCA 67.7 35.2 32.5 [176]
Calcite *** Wicking 57.0 40.2 16.8 1.3 54.4 [177]
GCC Wicking 37.0 29.1 7.9 0.5 31.6 [177]
GCC# Wicking 31.8 26.7 15.1 2.0 28.4 [177]
SBA DCA 46.44 31.38 15.06 [176]
SBA ** DCA 44.55 33.31 11.25 [176]
— 57 —
Table 3.3 - Surface energy of model sorbents as reported in the literature (continued)
Solid Method γ
T γD γ
P γ+ γ
- Ref.
mN/m mN/m mN/m mN/m mN/m
SB Tg=7°C
IGC 55.1 38.2 16.9 [178]
Note: PS, MCC, AC , GvOC, IGC, DCA, PCC, SBA, and GCC stand for polystyrene, microcrystalline cellulose, amorphous cellulose, Good-van Oss-Chaudhury, inverse gas chromatography, dynamic contact angle, precipitated calcium carbonate, styrene–butadiene–acrylonitrile latex and ground calcium carbonate; respectively.*Same solid measured with different set of liquid probes.** Washed.***Crystal.# after exposure to water vapour for 24h.
As observed in Table 3.3 there is variability in the surface energy values
obtained from the literature, especially in the case of regenerated cellulose and
calcium carbonate. These variations are due to differences in the methods
(contact angle, adsorption, thin layer wicking), the measurement conditions
(temperature, surface preparation, equilibration atmosphere, packing), the
theoretical treatment of the data (Zisman, Fowkes, GvOC, Owens and Wendt,
etc.) and the nature (roughness, AC film vs AC yarn, degree of grinding,
swellability) of the solid surfaces evaluated. The surface energy values extracted
from Table 3.3 and used in Chapter 6 of this thesis satisfied the following
criteria: closeness of the solid material, closeness of experimental conditions,
obtained using the preferred method for the solid and availability of polar and
dispersive components of the surface energy.
3.2.2.3.2 – Surface energy of treated fibers
There are two lines of thought for surface energy calculations based on contact
angle measurements: the surface tension component approach (Fowkes [179],
Owes-Wendt [180], van Oss-Good-Chaudhury [181]) and the equation of state
approach (Neumann-Spelt [182]) considered as a development of the Zisman
approach [183]. There still exists a controversy about which approach is better
[184]. In this study, due to the widespread use of the surface tension component
approach for paper and cellulosic fibers substrates [185], the method of Good-
van Oss-Chaudhury [186] was chosen to determine the surface energy of
— 58 —
handsheets produced with untreated and internally sized fibers. The method
requires contact angle measurements of three probe liquids of known surface
tension on the solid. Dynamic contact angles were measured with a FibroDat
contact angle machine at Xerox Research Centre of Canada. The samples were
conditioned for at least 24h in a controlled room at 23°C and 50%RH (according
to T402 Tappi Standard) before conducting the experiments. Eight contact angle
vs. time profiles per sample produced by the spreading and penetration of 4µl
drops on the solid were measured in the same conditioned room. The average
contact angle corresponding to 0.02s was used in the calculations. Equations 3.1,
3.2 and 3.3 were solved simultaneously for the unknown dispersive (γD), polar
(γP), electron-acceptor (γ+) and electron-donor (γ-) components of the fiber
surface energy with known values of the same surface energy components for
liquids i=1,2,3.
1+cosθ! γi = 2 ,�γiDγ
D�1/2+ �γi
+γ-�1/2
+ -γi-γ
+.1/2/ [3.1]
Table 3.4 summarizes the properties of the test liquids used for the surface
energy determination.
Table 3.4 - Surface energy components of test liquids
i Liquid γ
T
mN/m
γD
mN/m
γP
mN/m
γ+
mN/m
γ-
mN/m
Ref.
1 Diiodomethane 50.8 50.8 0 0.01 0 [187]
2 Formamide 58 39 19 2.28 39.6 [187]
3 Water 72.8 21.8 51 25.5 25.5 [187]
— 59 —
3.2.2.3.3 – Zeta Potential of model sorbents and fibers
The zeta potential of a few milligrams of each model sorbent and treated fibers
suspended in either distilled water pH 5 (low ionic strength) or 40mM potassium
phosphate buffer pH 6.8 (high ionic strength) was tested in a ZetaPlus zeta
potential analyzer (Brookhaven Instruments Corporation). The measurements
were conducted at 25°C using three independent samples and 10 runs per cycle.
3.2.2.3.4 – Scanning electron microscopy (SEM) of model sorbents and fibers
SEM micrographs of dry samples of the model sorbents and the fibers were
acquired in a JEOL JSM-840 microscope at different levels of magnification.
Except for the SB latex, all the samples were sputtered with carbon before
imaging.
3.2.2.3.5 – Specific surface area of model sorbents and fibers
The specific surface area of the sorbents was determined using the following
methods:
Table 3.5- Methods used for evaluation of specific surface area
Sorbent Method
Microgranular cellulose N2 and Methylene blue adsorption
Calcium carbonate N2 adsorption*
Rayon Methylene blue adsorption
Oxidized rayon Methylene blue adsorption
Polystyrene beads Calculation based on geometry
SB latex Methylene blue adsorption
Untreated fiber (BKSW) Methylene blue adsorption
Oxidized fiber (BKSW) Methylene blue adsorption
Note: * the data was informed by the supplier
— 60 —
A Coulter SA 3100 analyzer was used to measure the specific surface area by N2
gas adsorption followed by BET (Brunauer, Emmett and Teller) model fitting to
the experimental data. The methylene blue adsorption technique is based on the
measurement of the equilibrium isotherm of methylene blue adsorbed on the test
material followed by fitting of a Langmuir type of isotherm to the experimental
data [188]. The cross-sectional area of the adsorbed methylene blue dye was
assumed constant and equal to 197.2A2. This assumption can be questionable
depending on the orientation of the adsorbed molecule (end-on vs. side-on) and
the methylene blue concentration range (methylene blue forms dimmers and
trimers at concentrations above 10-5M). Nevertheless, the advantage offered by
the methylene blue method is the possibility of characterizing the specific
surface area of the materials under wet conditions. The method is particularly
useful in the case of fibrous materials that experience considerable amount of
swelling in aqueous media. Finally, for the spherical polystyrene beads with a
narrow particle size distribution, the specific surface area was calculated
assuming spherical particles with a uniform diameter, using the equation:
SSA = 6
ρs × d �0. ��
Where, SSA is the specific surface area [m2/g], ρs is the density of the solid sphere [g/cm3]
and d is the mean diameter [µm]
3.2.2.3.6 – Carboxylate content of oxidized cellulosic sorbents
The carboxylate content of the oxidized ground rayon and BKSW cellulosic
fiber was determined by simultaneous conductometric and potentiometric
titration. Initially, 0.3 g (dry basis) of sample was stirred in a solution containing
200ml of deionized water and 18 ml of 0.1M NaCl until the fibers were well
dispersed. The pH of the mixture was adjusted to fall within the range of 2.5 to 3
by adding 0.1M HCl. Then, the sample was titrated with 0.2ml increments of
— 61 —
0.04M NaOH until reaching pH 10. The evolution of both the conductivity and
pH of the mixture during the titration was recorded. From the plot of
conductivity vs. titrant volume, two end points can be identified: the first
corresponding to the equivalence point for the strong acid in the system (HCl)
and the second corresponding to the equivalence point for the weak acid in the
system (-COOH). The carboxylate content in the sample can be calculated from
the difference between these two volumes, as follows:
C = 0.04 × × × × 3333V2-V14444 w⁄⁄⁄⁄ [3.3]
Where C is the carboxylate content of the sample in [meq/g], V2 is the titrant
volume corresponding to the second equivalence point in [ml], V1 is the titrant
volume corresponding to the first equivalence point in [ml], and w is the weight
of sample (dry basis).
3.2.2.3.7 – X-ray photoelectron spectroscopy of the handsheets
X-ray photoelectron spectra of the increasingly sized handsheets (section
3.2.1.2) were acquired in a Thermo Scientific Theta Probe spectrometer using a
monochromated aluminum K-alpha radiation source (300µm spot size). Surface
charging was compensated with both low energy electrons and ions (Ar). The
vacuum level of the sampling chamber was maintained at 1×10-7mBar, with the
vast majority of residual chamber pressure due to residual Ar from operation of
the charge compensation source. Wide spectra (pass energy 200 eV) and spectra
of individual photoelectron lines C1s and O1s (pass energy 30 eV) were
acquired. The measured XPS spectra were not smoothed prior to deconvolution.
Peak fitting was carried out for high resolution C1 using the curve fitting
AVANTAGE software with the Gaussian-Lorentzian ratio of 70%/30%. The
binding energy scale was referenced to the C1s line of aliphatic carbon set at
285.0 eV.
— 62 —
3.3-BIO-INK PRINTING
3.3.1-Inkjet printer
A material deposition system based on piezoelectric inkjet technology (Dimatix
DMP 2800, see Figure 3.1) located inside a room with controlled atmospheric
conditions (23 ± 1°C temperature and 50 ± 2% relative humidity) according to
the T402 Tappi Standard was used. The printer includes a MEMS-based
cartridge-style disposable printhead with 16 nozzles linearly spaced at 254µm
and a typical drop size of 10pl. Each cartridge has a reservoir capacity of 1.5 ml.
The cartridge reservoirs were cleaned with both deionized water and buffer
before filling with ink. To avoid clogging of the printhead nozzles, all the
liquids were pre-filtered through a 0.2µm pore size Acrodisc Syringe Filter
(hydrophilic polypropylene membrane with low protein binding and low levels
of UV-absorbing extractables).
Figure 3.1(a) - Inkjet material deposition system
Figure 3.1(b) - Cartridge and print-head
(Images adapted from http://www.dimatix.com)
— 63 —
Figure 3.1c – Printhead nozzles: 16 nozzles, 254µm spacing, 21.5µm diameter.
The printer is equipped with a drop imaging system (drop watcher) that allows
observation and capture of the events during drop formation on the printhead
nozzles and trajectory of the drops after ejection. The drop imaging system is
based on bright background illumination with a stroboscopic LED array against
a ground glass screen that is synchronized with a monochrome CCD high speed
camera. The material deposition system also enables the user to control several
jetting parameters such as waveform, pulse width, individual nozzle voltage,
meniscus vacuum, firing frequency, cartridge temperature, and cleaning cycles.
3.3.2 - Ink-material compatibility
As advised by the supplier, the materials that get in physical contact with the
bio-ink during jetting are polypropylene (fluid module), peroxide treated EPDM,
ethylene propylene diene M-class rubber (housing of print module), and
silicon/silicon oxide (MEMS print chip). These materials were contrasted
against the bio-ink components to identify potential interferences or inhibitors of
the bioactive material and corrosion issues. The only issue identified was the
possibility of some residual peroxide in the EPDM that can interfere with the
— 64 —
activity measurements after printing. Bio-ink was jetted, collected after printing
and tested for peroxide. The bio-ink tested negative for peroxide after printing,
ruling out the presence of material incompatibility.
3.3.3 - Control of Jetting Performance
For a given bio-ink formulation and printing condition, visualization of the drop
formation process (see section 3.3.1 for details) at the nozzles allowed
evaluation of the jetting performance and troubleshooting of the corrective
actions needed to improve reliability. The strategy followed in the control of
jetting performance is presented in Figure 3.2.
Figure 3.2 – Measurement and control of jetting performance
3.3.4-Jettability test
The steps followed to test the jettability of a bio-ink formulation are described
below:
— 65 —
1. Run a few cleaning cycles (combination of priming, purging and spitting)
with the printhead located on the cleaning pad.
2. Set the jetting parameters to: maximum allowed nozzle voltage (40V), low
firing frequency (1 to 5 KHz), and widest pulse width in the waveform
compatible with the selected firing frequency.
3. Fire one nozzle at a time and groups of 5 nozzles at a time until the 16
nozzles are checked. Observe if the nozzles fire drops. If most of the nozzles
fire drops, proceed to step 7.
4. If most of the nozzles do not fire drops, modify the pulse width and observe
if there is drop ejection.
5. If most of the nozzles continue to not fire drops, change the cartridge. If the
nozzles start to fire, it is possible that there was a manufacturing defect in the
cartridge.
6. If most of the nozzles continue to not fire drops, the ink formulation is not
jettable.
7. Keep the nozzles firing for 10 minutes. If the nozzles do not fail to fire
during this time, proceed to step 10.
8. If within the 10 minutes, some or all of the nozzles fail to fire drops, modify
the pulse width and observe if there is some condition at which there is
sustainable firing.
9. If most of the nozzles continue to not fire drops, the ink formulation is not
jettable.
10. Determine the drop speed in m/s, using the strobe delay and the distance
travelled by the drops.
11. Decrease the pulse width in small steps and determine the drop speed for
each operational condition.
12. Determine the optimum operational window (pulse widths) corresponding to
the speeds 7 to 9 m/s suggested as optimum by the manufacturer.
13. Decrease the firing voltage until no drops are jetted, and select a voltage
higher than this limit as the operational firing voltage.
14. Increase the firing frequency and repeat the procedure.
— 66 —
15. The final results of the procedure are the optimum pulse width, firing voltage
and firing frequency for the ink under evaluation.
16. With the optimum operational parameters and the same cartridge, print once
per day during one week. Repeat steps 1 and 3; if firing problems are
observed is possible that the ink under evaluation produces a “first drop
problem”.
17. If firing problems are not observed, the ink formulation is a good candidate
to proceed to ink stability tests.
3.3.5-Printed patterns
Three types of patterns were printed, as detailed in the sketch of Figure 3.3
Figure 3.3 – Sketch (not to scale) of inkjet printed patterns. Spot (left), lines (center) and
dots (right)
3.4-ENZYME MEASUREMENTS
3.4.1-Activity in solution
One unit of HRP enzyme activity is defined as the volume of peroxidase enzyme that will
oxidize 1µmol of ABTS per minute at pH 6.8 and 25°C. Therefore, HRP enzymatic
activity was tested in solution through absorbance measurements of the concentration of
oxidized ABTS ―a green colored chromogen― over time using a Perkin Elmer UV-Vis
spectrometer Lambda 35 operated at ambient temperature (22 with maximum/minimum
of ± 3°C) and 422 nm. Activity determinations were performed in triplicate
20µm
20µm
20µm
254µm 254µm
254µm
— 67 —
measurements of the same sample, following the method described in [ 189 ] and
reproduced in Appendix A.1 using concentrations of 1.68nM HRP, 1.21mM ABTS and
0.13mM H2O2 in the corresponding liquid carrier. The enzyme activity was calculated
using the maximum linear rate (initial slope) of the absorbance curve and applying the
following equation:
A = S × V × df
6 × 78 �0. 9�
A activity [U/ml]
S slope [a.u./min]
V total volume [ml]
df dilution factor
εεεε extinction factor [mM-1cm-1]
Ve enzyme solution volume [ml]
Standard errors as high as ±10% are typical in HRP activity measurements in
solution using ABTS as chromogenic substrate [190,191,192]. Temperature
changes were the most significant source of variations in activity measurements
(see Appendix A.5). At least a 5% change in activity measurements can be
explained by temperature changes.
3.4.2-Activity in solution (after printing)
In order to determine HRP activity after inkjet printing, a minimum of 10
million drops (equivalent to 100µl) of the HRP ink was jetted using the Dimatix
printer onto a small plastic cup. Three replicates of the post-jetted HRP ink
(50µl each in volume) were used to measure enzyme activity in solution
following a similar procedure as described in section 3.4.1.
3.4.3 – Protein concentration in solution
Protein concentration in solution was measured following Bradford protocol
[193]. Bradford dye can exist in three forms, depending on the pH:
— 68 —
(Alkaline) (Neutral) (Acid)
Anionic ↔ Neutral ↔ Cationic [3.5]
595 nm (blue) 650 nm (green) 470 nm (red)
The technique involves the formation of a dye-protein complex that shifts the
absorbance peak from red to blue. Detergents, bases, and other substances can
produce interferences [194]; therefore, a suitable blank containing all the sample
components including potential interfering substances except the protein were
subtracted from the sample absorbance measurements.
The protein concentration in solution before and after adsorption was measured
in duplicate following the Bradford microassay protocol for protein
concentrations below 0.1 mg/ml. For the higher protein concentrations, dilutions
with supernatant were implemented. In both cases, a protein calibration curve
was constructed using HRP solutions with known enzyme concentrations.
3.5- SOLID PHASE BIOANALYSIS OF H2O2
3.5.1-Principle
Bioanalysis of H2O2 is based on the reaction between the leuco-dye, 2,2’-azino-di-(3-
ethylbenzthiazoline) 6-sulphonate (ABTS) and H2O2 specifically catalyzed by the
enzyme HRP [195] as follows:
HRP
2ABTS + H2O2 →→→→ 2ABTS + 2H2O [3.6]
(reduced) (oxidized)
In the reduced form, the chromogen is colorless. As the enzymatic reaction
proceeds, a green color corresponding to the oxidized form of ABTS is
developed in direct proportionality to the amount of H2O2 converted.
— 69 —
3.5.2-Color development
For the quantitative determination of H2O2, the backside of the printed paper
strips was kept in contact with solutions of H2O2 of known concentration until
the solution fully penetrated the paper thickness (see Figure 3.4). The excess
solution was removed with a blotting paper. The paper samples were left to dry
in air in the conditioned room (23 ± 1°C temperature and 50 ± 2% relative
humidity, according to the standard Tappi T402) and unless stated differently,
the color response was measured after 20min and within one hour of having
started the reaction.
a – Initial paper b – Bio-ink inkjet printing c – Contact with solution
d – Removal of excess e – Air drying and color development
Figure 3.4 – Schematic illustration of H2O2 bioanalysis using paper
3.5.3-Measurement of color response
The diffuse reflectance spectrum of the dry printed spots was measured for the
wavelength interval 400 to 700 nm using a color reflection spectrodensitometer
upper side
back side
upper side
back side
printed spot H2O2 solution
blotting paper
color development
— 70 —
X-Rite 530 within 1 hour of exposure to the H2O2 solution. The illuminant D65
and the observer geometry 2° was selected for the measurements. Also, a
flexible thin film UV filter that allowed less than 10% transmission below 390
nm was used on top of the samples during the color measurements to remove the
contribution from optical brighteners present in some of the commercial paper
samples. All the measurements were performed in the conditioned room
described before.
Two values were extracted from the spectrum: the wavelength corresponding to
maximum reflectance (to characterize the color profile) and the magnitude of the
reflectance at the previous wavelength (to characterize the color intensity). The
optical measurements were taken at two points in time: a) after 20 min and
within 1 hour of exposure to the H2O2 solution and b) after 1 hour of exposure to
the H2O2 solution. Previous experiments (see Appendix A.2) have found that
before 20 min, the color intensity, i.e. the magnitude of the reflectance, changes
over time due to the progress of the reaction and the paper drying process, and
after 20 min, the color intensity reaches a steady state.
3.5.4-Bioanalysis calibration
Kubelka Munk theory [196] can be used to linearly relate the diffuse reflectance,
R, measured on the paper color spots and the analyte concentration, C, making
quantitative analysis possible. Following a similar approach as in [150], the
function AR = -ln(R/Rb) vs. log C was plotted as a calibration curve. AR
represents the solid phase analogous of absorbance in liquid media, R is the
diffuse reflectance of the color spot for an infinite thick layer of paper and Rb is
the diffuse reflectance of the blank for an infinite thick layer of paper. A printed
spot exposed to deionized water was used as a blank.
— 71 —
3.6-ENZYME SPATIAL DISTRIBUTION
3.6.1 – Principle
Fluorescent detection of H2O2 is based on the reaction between a fluorescent probe, 10-
acetyl-3,7-dihydroxyphenoxazine, Amplex Red (supplied by Molecular Probes) and H2O2
as follows:
HRP
Amplex Red + H2O2 →→→→ Resorufin + 2H2O [3.7]
(non-fluorescent) (red fluorescent)
HRP catalyzes the reaction of Amplex Red (non fluorescent) with H2O2 to produce
resorufin, a red-fluorescent oxidation product with a 571nm absorption peak and a 585nm
fluorescence emission maximum (See Appendix A.3). The reaction has been used to
detect concentrations of peroxidase in solution as low as 1 x 10-5 U/ml [197].
3.6.2 – New bio-ink formulation
Amplex Red was initially dissolved in dimethylsulfoxide (DMSO) according to the
supplier instructions to obtain a 10 mM solution. A 50 U/ml (pyrogallol units) HRP
enzyme and 0.12 mM Amplex Red bio-ink formulation was prepared in a 40mM KH2PO4
buffer (pH 6.8) containing 0.1 wt-% Triton X-100 as a surfactant, 10 wt-% glycerol as a
humectant and a 0.5 wt-% carboxymethyl cellulose as viscosity modifier.
3.6.3 – Inkjet printing and fluorescence development
The new bio-ink formulation was printed on the paper supports under the same
conditions detailed before. The printed papers were exposed to a 0.1mM H2O2 solution to
develop the fluorescent response. Positive and negative controls were also prepared. A
sample completely soaked in the Amplex Red-containing bio-ink was used as the positive
control. A sample completely soaked in the Amplex Red-containing bio-ink free from the
HRP enzyme was used as the negative control.
— 72 —
3.6.4 – Embedding and sectioning
Part of the samples resulting from the previous step (section 3.6.3) were embedded using
SPI-Pon™ 812 Epoxy Embedding Kit (SPI microscopy supplies, Canada), cured at 60°C
for 24h and cross-sectioned with a diamond knife in a Leica Ultramicrotome operated at
room temperature. The sections were discarded and the blocks were kept for microscopy.
3.6.5 – Confocal laser scanning microscopy (CLSM)
Either the printed samples (surface view) or the blocks of the embedded printed samples
(cross-sectional view) were imaged with a two-detection-channel laser confocal
microscope, Leica TCS SP2. The surface views were obtained under 5X or 20X air
objectives; whereas, the cross-sectional views were obtained under a 63X oil immersion
objective (HC PL APO CS, NA 1.4). The microscopy conditions are detailed in Table 3.3.
Table 3.6- Laser confocal microscopy conditions
Excitation Beam Splitter Channel Emission/Reflection Image
Green HeNe Laser
100% 543nm
DD 488/543
1 Red Visible Fluorescence
555nm -700nm
Active HRP enzyme map
2 Reflection 540nm-545nm
High scattering power particles
(pigments/fillers)
3.7 – ADSORPTION ISOTHERMS
3.7.1 – Enzyme solutions
Solutions of HRP in 40mM KH2PO4 buffer of pH 6.8 with concentrations
ranging between 0 and 0.4 mg/ml were used in adsorption experiments.
— 73 —
3.7.2 - Depletion method
The HRP solutions with concentrations ranging between 0 and 0.4 mg/ml were
separated in three parts; A, B and C. Part A was used to determine a protein
calibration curve (according to the procedure detailed in section 3.4.3). Part B
and part C were defined as control and sample, respectively. The samples
consisted of mixtures of model sorbent and enzyme solution (part C) in the
ratio of 0.7 m2 of sorbent per 5ml of enzyme solution. The control consisted of
5ml of enzyme solution (part B) with no sorbent added. Both the control and
the sample were transferred to centrifuge tubes and were left at room
temperature in a horizontal shaker for 24h. After adsorption, the control and
sample tubes were centrifuged for 30min, at 4°C and 5000rpm (except for
polystyrene beads that were centrifuged at 10000rpm). Next, protein
concentration and enzyme activity were tested in the supernatant.
3.7.3 – Adsorption isotherm construction
Adsorption isotherms represent the amount of enzyme adsorbed per unit area of
sorbent material as a function of the equilibrium enzyme concentration in
solution. The protein concentration in the supernatant after 24h adsorption
(sample tube) was taken as the equilibrium enzyme concentration. The
difference (depletion) between protein concentration in the supernatant of the
control and the sample tubes after 24h adsorption normalized by the test volume
and the area of sorbent per tube was used to calculate the amounts of enzyme
adsorbed:
Γ = Ce - Co !!!! × Vt
SSA × w × 1000 ����3.8����
Where:
Γ adsorbed amount [g/m2]
Ce protein concentration in sample tube supernatant [mg/ml]
Co protein concentration in control tube [mg/ml]
— 74 —
Vt total volume in tube [ml]
SSA specific surface area of the sorbent [m2/g]
w weight of sorbent in tube [g]
3.7.4 – Inactivation isotherm construction
In the context of this thesis, inactivation isotherms represent the percentage change in
enzyme activity free in solution upon adsorption as a function of the equilibrium enzyme
concentration in solution. The protein concentration in the supernatant after 24h
adsorption (sample tube) was taken as the equilibrium enzyme concentration. The
percentage difference (inactivation) between HRP activity in the supernatant of both
control and sample tubes after 24h adsorption was calculated according to:
I = Ae - Ao !!!! × 100
Ao ����3.9����
Where:
I inactivation [%]
Ae HRP activity in sample tube supernatant [U/ml]
Ao HRP activity in control tube supernatant [U/ml]
3.7.5 – Modeling of adsorption isotherms
The models described below, were fitted to the experimental adsorption data
using a nonlinear regression package (SigmaPlot 2001, version 7.101) based on
the Levenberg-Marquardt algorithm.
3.7.5.1 - Langmuir’s modeling
The model used is described in equation 3.9:
: = :;<= > ?8 ! � + > ?8 !⁄ �0. �@�
— 75 —
Where:
Γ adsorbed amount [µmol/g]
Ce equilibrium protein concentration [µmol/l]
K binding affinity constant [µmol/l]-1
Γmax adsorption capacity of the sorbent [µmol/g]
3.7.5.2 - Freundlich’s modeling
The model used is described in equation 3.10:
: = > ?8; �0. ��� Where:
Γ adsorbed amount [µmol/g]
K constant [µmol(1-m).g-1.lm]
Ce equilibrium protein concentration [µmol/l]
m heterogeneity index
3.8-STABILITY MEASUREMENTS
3.8.1-Bio-ink storage stability
A test solution of standard bio-ink formulation (See section 3.1.6) and a control
solution containing HRP in a 40mM KH2PO4 buffer with pH 6.8 were kept in a
refrigerator at 4°C. At defined time intervals, HRP activity was tested on aliquot
portions of the test and control solutions following the procedure described in
section 3.4.1. The percentage change in HRP activity with respect to the control
solution was calculated.
3.8.2-Bioactive paper storage stability
Uncoated wood-free paper was printed with the standard bio-ink and was kept in
dark sealed envelopes under three storage conditions: freezer (-20°C),
— 76 —
refrigerator (4°C), and a conditioned room for paper testing (23°C, 50%RH).
The bioanalytical response of the papers was evaluated over time for low
(0.3mM), intermediate (1.5mM), and high concentrations (3mM) of H2O2,
following the steps detailed in section 3.5.
3.8.3-Adsorbed HRP thermal stability
A differential scanning calorimetry (DSC) technique was used to characterize
the thermal unfolding of a 10 mg/ml HRP solution in 40mM KH2PO4 buffer pH
6.8 with and without purification by dialysis against the buffer, and HRP
adsorbed on increasingly sized fibers (corresponding to 0.4mg/l HRP initial
solution). The fiber samples with adsorbed HRP were taken from the pellets left
after removal of liquid by centrifugation during adsorption experiments (See
section 3.7.2). The thermograms were obtained in a Q1000 DSC calorimeter
(TA Instruments) using aluminum high volume pans. The measurement
temperature range was 10°C to 150°C with a 1°C/min scanning temperature
ramp. A empty aluminium high volume pan was used as the reference and the
thermogram corresponding to a second temperature ramp on the same sample
was used as the baseline for correction. All curves were smoothed with a 6-point
moving average to remove noise. The weight of the sealed pans containing the
samples was determined before and after thermal scanning to verify hermeticity.
The thermograms corresponding to runs with significant weight variations were
disregarded.
— 77 —
Chapter 4
INK FORMULATION AND PIEZOELECTRIC INKJET PRINTING
4.1-INTRODUCTION
Several additives are incorporated in inkjet formulations to adjust the physicochemical
properties of the inks and make them stable and ejectable [80,81,198,199,200]. Inkjet
printers with different actuation strategies have different requirements for ink properties.
In the case of drop on demand systems, DOD, piezoelectric inkjet requires a higher ink
viscosity (5-10cps) than thermal inkjet (1-1.5cps). In a typical ink formulation of
commercial piezoelectric inkjet for graphic arts applications, a number of polyhydric
alcohols such as glycols, glycerol, and diols are used as viscosity modifiers [84, 89].
However, if bioagents, such as enzymes, are to be inkjet printed, an additional factor
must be considered in the ink formulation: preservation of the biological functionality
and stability of the bio-ink over time [6]. It is well known that bioagents, such as
enzymes exhibit different degrees of denaturation if physical conditions such as moisture,
temperature, pressure or mechanical stress, and chemical factors like pH, presence of
denaturing substances (alcohols, heavy metals, detergents), and ionic strength are not
kept at optimum values [10]. Thus, inappropriate ink additives, operation conditions, and
viscosity modifiers will negatively affect the enzyme activity. Additives used to optimize
ink rheological parameters may produce inactivation or denaturation of the enzyme
immediately or over time.
A suitable bio-ink formulation has to be able to maintain the activity of the enzyme and at
the same time produce stable and repeatable drops during jetting. In order to jet the bio-
ink, the viscosity and surface tension of the bio-ink (initially in the range 60-65 mN/m
and 1-1.2 cps, respectively, without additives) had to be adjusted to the optimum values
(30 mN/m and 5 cps) as suggested by the literature and the particular printhead
requirements [78, 201,202].
— 78 —
Table 4.1 compares a generic commercial ink formulation [80,203] with a potential
design of enzyme bio-ink.
Table 4.1 – Comparison between a generic commercial ink and enzyme bio-ink
formulations
Component Commercial Ink
Enzyme Bio-ink
Functional Objective
Main ingredient
Pigment/dye X Coloring
Enzyme X Biorecognition
Chromophore X Colorimetric transduction
Vehicle
Solvent X X Diluent and control of ink drying
Co-solvent X
Buffer X X Control of pH
Additives
Surfactant X X Control of surface tension
Viscosity modifier X X Control of viscosity
Biocide X Prevent biological growth
Chelating agent X Avoid deposit formation
Dispersant X Prevent agglomeration
Protein stabilizer X Increase enzyme stability
Humectant X X Prevent nozzle crusting
— 79 —
4.2-SPECIFIC OBJECTIVE
The first goal of this thesis chapter is to systematically study the impact of piezoelectric
ink additives, in particular viscosity modifiers, on HRP enzymatic activity. To achieve
this objective, ink formulations containing various concentrations of different viscosity
modifiers and a non ionic surfactant are evaluated (see Table 3.1 in Chapter 3).
The second goal of this thesis chapter is to systematically study the jetting performance
as a function of ink formulation and adjustable jetting parameters. To achieve this
objective the bio-ink formulations found to produce minimal enzyme inactivation were
jetted using a piezoelectric research printer run under different operational conditions.
The ink that produced the most reliable jetting performance was selected as the standard
bio-ink formulation. Subsequently, the impact of the jetting process on enzyme activity
was investigated.
A version of this chapter was published in Macromolecular Rapid Communications,
March 2007 [204].
4.3 - RESULTS
4.3.1 - Control of bio-ink surface tension
To adjust the bio-ink surface tension, Triton X-100 ((Trademark, Rohm and Haas Co.,
Philadelphia, iso-octyl phenol monoethylene glycol ether), a non ionic surfactant, was
used instead of anionic and cationic surfactants due to its lower impact on enzyme
activity [205]. Initially, buffer solutions with increasing concentrations of surfactant were
prepared and their surface tension was measured to determine the critical micelle
concentration and the minimum surface tension achievable. Figure 4.1 illustrates the
results obtained. The critical micelle concentration (cmc) for the system studied was 0.05
wt.-% of surfactant concentration. For surfactant concentrations above the cmc, surface
tension stabilizes in its lower limit: approximately 30mN/m at room temperature. This
lower surface tension value is in agreement with values reported in the literature [206]
and is within the optimum range for piezoelectric inkjet printing.
— 80 —
Figure 4.1 - Effect of Triton X-100 dosage on surface tension of 40mM potassium
phosphate buffer at pH 6.8 – The error bars represent the ± one standard deviation for 6
replicate surface tension measurements of the same sample. The maximum observed
value of the standard deviation was 0.8mN/m.
Because enzymes can become denatured in the presence of detergents, solutions of HRP
0.05µM in buffer containing various concentrations of surfactant were prepared. HRP
activity in these solutions was measured immediately after preparation and contrasted
with the HRP activity of a control solution of similar concentration without surfactant.
Figure 4.2 shows the impact of the surfactant dose on HRP activity. It is clear that Triton
X-100 is a mild detergent for HRP; no significant impact of the surfactant on HRP
activity is detected within experimental error (±10%).
Solutions containing buffer and a number of viscosity modifiers (in various
concentrations) were formulated with a fixed 0.1wt.-% surfactant concentration. Surface
25
30
35
40
45
50
55
10-5 10-4 10-3 10-2 10-1 100
Triton X-100 dose (wt.-%)
Surf
ace
tens
ion
(mN
/m)
— 81 —
tension of all the solutions at room temperature (at the optimum viscosity of 5cps) was
found to be around 30 mN/m, which is the optimum value required by the printer.
Figure 4.2 - Effect of Triton X-100 dosage on activity of 0.05µM HRP in 40mM
potassium phosphate buffer at pH 6.8. Three replicate activity measurements of both the
sample and the control were performed. The change in activity was calculated with the
average activity values corresponding to the sample and the control. Variations of ±10%
are within the experimental error of the HRP activity assay (See section 3.4.1).
4.3.2 - Control of bio-ink viscosity
To adjust the bio-ink viscosity, some viscosity modifiers typically used in commercial
ink formulations and some common enzyme stabilizers were explored [84]. The viscosity
modifiers and the range of concentrations investigated were detailed in Chapter 3, Table
3.1. Figure 4.3 summarizes the impact of different viscosity modifiers on the room
temperature viscosity of a solution containing the buffer and 0.1wt.-% of the surfactant.
All solutions exhibited an exponential dependence of the viscosity on the concentration
-30
-20
-10
0
10
20
30C
hang
e in
Act
ivit
y (%
)
Surfactant dose (wt-%)10-5 10-4 10-3 10-2 10-1 100
— 82 —
with different sensitivities depending on the type of the viscosity modifier. For the same
type of viscosity modifier (for instance PEG), lower concentrations of the higher
molecular weight modifier were needed to reach a target viscosity value, as expected.
Figure 4.3 - Effect of the dose of different viscosity modifiers on the ambient
temperature viscosity of solutions containing 40mM potassium phosphate buffer pH 6.8
and 0.1 wt.-% Triton X-100. The error bars represent the ± one standard deviation for 3
replicate viscosity measurements of the same sample. The maximum observed value of
the standard deviation was 0.2 cps (corresponding to 0.01 in logarithmic viscosity scale).
Even though some researchers have reported activity gains when low concentrations of
polymers (such as polyethylene glycol), globular proteins (such as hemoglobin and
lysozyme), or carbohydrates (such as dextrans) were incorporated in some enzymatic
systems, significant losses in enzyme activity have been observed in the presence of high
concentrations of high molecular weight components [207, 208]. The losses in activity
were explained in terms of the decrease in rate of diffusion of the substrate and products
to and from the enzyme active sites and the decreased motion during conformational
changes in motile enzymes [ 209]. The viscosity modifiers investigated in our study
10-1
100
101
102
103
0% 20% 40% 60% 80%
Viscosity modifier dose (wt.-%)
Vis
cosi
ty(c
ps)
PEG 20,000 PEG 2000PEG 200 GlycerolEthylene glycol PVACMC
— 83 —
contained high molecular weight substances, and in some cases (for instance glycerol and
ethylene glycol), very high concentrations were required to increase the ink viscosity to
the target value (5cps).
Therefore, the activity of the bio-inks formulated with 0.05µM HRP in 40mM potassium
phosphate buffer with 0.1wt.-% Triton X-100 containing different concentrations of
seven types of viscosity modifiers was evaluated at room temperature immediately after
preparation. In Figure 4.4, the percent change in HRP activity resulting from the addition
of the viscosity modifiers is presented. Negative values indicate reduction in activity with
respect to the control solution (without viscosity modifier) and vice versa.
Figure 4.4 - Impact of viscosity on HRP activity for a solution containing 0.05 µM HRP
in 40mM potassium phosphate buffer at pH 6.8 with 0.1 wt.-% Triton X-100 and variable
doses of the viscosity modifiers indicated in the plot legend. Three replicate activity
measurements of both the sample and the control were performed. The change in activity
was calculated with the average activity values corresponding to the sample and the
control. Variations of ±10% are within the experimental error of the HRP activity assay
(See section 3.4.1).
— 84 —
From Figure 4.4, with the exception of CMC, significant decreases in HRP activity were
observed when viscosity modifiers were added to the enzyme carrier. Similar results were
found by Derham and Harding [207] in their study of the effect of elongated polymers on
urease activity. They claimed that when enzymes existing in oligomeric states are placed
in a medium exhibiting macromolecular crowding (i.e., a significant proportion of the
volume available for the enzyme is occupied by other molecules), increased self-
association may result in enhanced activity. However, if the concentration of the
macromolecules (other than the enzyme) in the solution is too high, limitations in
diffusion can decrease the reaction rate.
Moreover, from Figure 4.4 it can be observed that for the PEG series, the higher the
molecular weight, a more pronounced decrease in HRP activity is found. Most likely this
is related to lower diffusion coefficients of the substrate and products to and from the
enzyme active sites due to the larger molecular size of the viscosity modifier. In addition,
despite having used a highly purified grade of PEG with less than 0.001% peroxides as
H2O2, some residual H2O2 was detected in blank solutions, creating interference for
activity determinations. If PEGs are to be used in HRP containing inks, they would
require prior removal of the residual H2O2.
Researchers have also found that the nature of the interactions between cosolvents (in this
case the viscosity modifiers) and proteins (in this case HRP) dictate if a cosolvent will act
as stabilizer or a deactivator. Co-solvents that are preferentially excluded from the
surface of proteins tend to favor the native folded state of the protein. Sugars and PEG at
low concentrations tend to show this behavior. Co-solvents that form stronger hydrogen
bonds than water (glycerol and polyols) also tend to be excluded from the protein surface;
hence, are stabilizers. However, co-solvents that can interact with non polar patches of
the protein surface through hydrophobic interactions tend to preferentially exclude water,
favoring the unfolded inactive state. PEG at high concentrations or with high molecular
weight interacts with proteins using mainly non polar chains; hence, is a destabilizer.
CMC appeared to be the most effective viscosity modifier. A significantly lower
concentration of CMC was required to achieve the desired range of viscosity and the
— 85 —
HRP activity in the resulting ink was not significantly changed. This behaviour may be
the result of the distinct molecular structure and physicochemical properties of CMC.
Unlike PEGs and PVA that are neutral polymers, CMC is a charged polymer. CMC’s
viscosity in solution not only depends on concentration, molecular weight, and flexibility,
but also on charge density [210]. This result implies that macromolecular overcrowding
was not present in the ink containing CMC.
4.3.3 – Bio-ink storage stability
The standard bio-ink formulation containing enzyme, buffer and additives was kept in a
fridge at 4°C and tested for HRP activity at defined time intervals to determine the bio-
ink storage stability. A solution containing only enzyme and buffer (free from additives)
was used as a control. Figure 4.5 illustrates the changes in activity for both control and
optimized formulation (sample) over time.
Under the storage conditions described above, it can be observed from Figure 4.5 that the
optimized enzyme formulation exhibits a similar behaviour as the control solution prior
to day 57. This result indicates that the ink additives do not significantly alter the storage
stability of the bio-ink. In addition, it can be assumed that the storage stability of the bio-
ink follows a biphasic behavior: remains constant for the first 40 days (phase 1) after
which significant losses of enzyme activity are observed (phase 2). The important
variability observed in the experimental data is consistent with similar studies reported by
the supplier [ 211] and might be explained by slight temperature changes during the
activity measurements (see section 3.4.1 and Appendices A4 and A5).
— 86 —
Figure 4.5 – Temporal evolution of peroxidase activity in the standard HRP bio-ink.
Control: 50U/ml HRP in PBS pH6.8; Sample: 50U/ml HRP in 40mM potassium
phosphate buffer at pH6.8 with 0.1wt.-% Triton X-100, 0.5wt.-% CMC, and 10wt.-%
glycerol. Activity measured in terms of change in absorbance per unit time. Error bars
represent the ±one standard deviation corresponding to three replicate measurements of
the same sample.
4.3.4 – Printing window for HRP bio-ink
The jettability of the HRP ink containing variable amounts of carboxymethyl
cellulose (CMC) as the viscosity modifier was investigated in the piezoelectric
material deposition system described in section 3.3.1. Several jetting
experiments were performed with variable firing frequency, pulse width,
reservoir pressure, and firing voltage. Following an approach similar to the one
reported by Kang [82], plots of drop speed vs. pulse width were obtained. Figure
4.6 shows velocity-pulse width profiles obtained for different concentrations of
CMC at a fixed firing frequency, driving voltage, and ink temperature.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0 10 20 30 40 50 60 70 80
Act
ivit
y (A
U/m
in)
Days
ControlSample
— 87 —
Figure 4.6- Velocity profiles of 0.05µM HRP in 40mM potassium phosphate
buffer pH6.8, 0.1wt.-% Triton X-100 and variable amounts of CMC. Firing
frequency is 3 KHz, driving voltage is 30V and ink firing temperature is 28°C.
Taking into consideration the optimum range of drop speeds suggested by the
printer manufacturer (7-9 m/s), it was found that the minimum pulse width that
produced stable monodisperse drops was approximately 70µs. This optimum
pulse width is large when compared with the typical pulse width values reported
in the literature for piezoelectric drive drop ejectors (5-20µs) using commercial
inks [82,84]. Nevertheless, in the literature [44] larger pulse widths are reported
for enzyme piezoelectric jetting (50µs, 200Hz). Thus, a lower ejection rate was
needed for the reliable jetting of the HRP inks. The bio-ink containing 0.2wt-%
CMC (low viscosity) produced drops too fast and a larger pulse width was
required to reduce the drop speed. The bio-ink formulations containing 0.5 wt-%
CMC and 0.8wt-% CMC resulted in a similar range of pulse width for optimal
drop speed. To keep the ink formulation simple and inexpensive and the enzyme
activity unaltered, the lower addition level of CMC was selected.
0
5
10
15
20
25
30
0 20 40 60 80 100 120 140
drop
vel
ocit
y (m
/s)
pulse width (us)
0.2 wt.-% CMC
0.5wt.-% CMC
0.8wt.-% CMC
— 88 —
Table 4.2 summarizes the range of variations of each jetting parameter, the
optimum values found and the impact of lower and higher settings on the
performance, Figure 4.7 illustrates the waveform shape that produced reliable
jetting of the bio-ink.
Table 4.2 - Jetting conditions
Parameter Available Range
Lower Optimum Higher
Nozzle Voltage 0-40 V No drops are ejected
30V Chaotic spray
Waveform Single or multiple pulses with variable fall/rise times
Single pulse Single pulse. Fall in two
steps
Double, multiple or
bipolar pulse
Pulse Width variable No drops are ejected
70 µs Drop speed is too low
Meniscus Vacuum
0-5 inches H2O Nozzle covered by fluid
4.5-5 Air ingestion
Firing Frequency
0-40kHz depending on pulse width
Time consuming
3kHz Limited by pulse width
In addition, cleaning cycles and non-jetting waveform were investigated.
Priming of the cartridge at the beginning of the printing job with a spit-purge-
spit multiple cleaning cycles helped eliminate trapped air from the ink circuit
and dry ink deposits from the nozzle entrance. During printing, a spit cleaning
cycle conducted every 5 bands (for bio-ink containing ABTS chromogen) and
every 10 bands (for bio-ink free of ABTS) helped keep the printing operation
stable. Also, the nozzles were excited with low amplitude periodic voltage
pulses (tickle) during non-jetting periods to prevent the formation of ink
deposits at the nozzles.
— 89 —
Figure 4.7- Driving waveform for reliable jetting of HRP bio-ink
Following the link below, a movie of the drop firing process for the bio-ink
containing 0.5wt-% CMC with optimized jetting conditions is provided as a
supplementary material. (See Appendix A.6 for further details).
CMC05.avi
[Supplementary material 4.1]
4.3.5 -First drop problem
An additional problem that needs to be addressed for reliable jetting is the first
drop problem. This problem is caused by the evaporation of solvent at the
nozzles during idle periods. The evaporation resulted in local changes in the ink
composition and rheological properties which required an increased driving
force for drop ejection. Potentially it can lead to nozzle clogging. To check for
the presence of this problem, the same ink and cartridge were repeatedly used
over a period of one week, once a day. During the idle periods, the cartridge
Firing
voltage
Vo
lta
ge
Time
Pulse width
Standby
Zero
70µs
30V
Firing
frequency
3kHz
— 90 —
was stored in a fridge at 4°C to preserve the enzyme activity. Absent or erratic
drop firing was observed, indicating the existence of the first drop problem. In
the presence of the problem, the immediate action taken was to increase the
nozzle voltage, to run a few cleaning cycles before starting to print and to
observe if there was improvement in the jetting performance. In some cases, this
action restored the optimum jetting characteristics; in other cases, this was
insufficient to produce drop firing. To reduce evaporation and to enhance ink
performance, 10% w/w of glycerol was added to the formulation as a humectant.
It was observed that the humectant did not affect the printing operation window;
the first drop problem was avoided and ink drops with approximately
10±1ng/drop were consistently jetted.
4.3.7 – Impact of jetting on HRP activity
The activity in solution of the HRP ink containing CMC as the viscosity
modifier was measured (see section 3.4.2) before and after jetting under the
optimum conditions described in the previous section. It was found that the
percentage change in HRP activity due to the jetting process was less than 2%,
and it was not statistically significant at a 5% confidence level. The slight
increase in enzyme activity after printing could have been associated with
sample evaporation.
— 91 —
Figure 4.8 – Comparison of enzyme activities in solution for different steps
during printing. Error bars represent the ±one standard deviation of four
independent ink samples. Each sample activity corresponds to the average of
three replicate measurements. Results of Student’s two-tail T-test indicated on
the plot.
4.3.8 – Bioactive paper storage stability
A pattern of square spots was printed on several sheets of uncoated wood-free paper
using the standard HRP-containing bio-ink. The enzyme-printed paper was kept inside
dark sealed envelopes under three storage conditions: freezer (-20°C), fridge (4°C) and
conditioned room for paper testing (23°C, 50%RH). After defined time intervals, paper
strips were removed from the envelopes and used to analyze solutions of low,
intermediate, and high H2O2 concentrations. Figures 4.9, 4.10 and 4.11 show the color
response (measured as AR) over a period of 8 months for the three different storage
conditions.
p = 0.8
p = 0.3
p < 0.05
0.0
0.1
0.2
0.3
0.4
Initial solution Solution in cartridge
Solution after printing
[A.U
./min
]
— 92 —
Figure 4.9 – Temporal evolution of the color response (measured as AR) of HRP-printed
paper stored in a freezer (-20°C) and exposed to solutions of low, intermediate, and high
H2O2 concentrations. Error bars represent the ±one standard deviation in the AR value
corresponding to the simultaneous color development of ten enzyme spots printed on the
same paper strip.
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0 25 50 75 100 125 150 175 200 225
AR
Days
0.1 mM
0.5 mM
1 mM
— 93 —
Figure 4.10 - Temporal evolution of the color response (measured as AR) of HRP-printed
paper stored in a fridge (4°C) and exposed to solutions of low, intermediate, and high
H2O2 concentrations. Error bars represent the ±one standard deviation in the AR value
corresponding to the simultaneous color development of ten enzyme spots printed on the
same paper strip.
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0 25 50 75 100 125 150 175 200 225
AR
Days
0.1 mM0.5 mM1 mM
— 94 —
Figure 4.11 - Temporal evolution of the color response (measured as AR) of HRP-printed
paper stored in a conditioned room for paper testing (23°C, 50%RH) and exposed to
solutions of low, intermediate, and high H2O2 concentrations. Error bars represent the
±one standard deviation in the AR value corresponding to the simultaneous color
development of ten enzyme spots printed on the same paper strip.
When stored in the freezer, the printed enzyme shows the same or better color response
overtime for at least 8 months within experimental errors (See Figure 4.9). When stored
in the conditioned room for paper testing, the color response of printed HRP is
maintained for at least 3 months after which the color response starts to decay (See
Figure 4.11). Finally, for the HRP-printed papers stored in the fridge there is a
statistically significant decrease in AR after 1 month (See Figure 4.10). It was also
observed (results not shown) that the initial color of the printed spots before bioanalysis
tended to become greener overtime in the order of conditioned room>fridge>freezer.
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0 25 50 75 100 125 150 175 200 225
AR
Days
0.1 mM0.5 mM1 mM
— 95 —
It is not clear why storage in the fridge resulted in the poorest immobilized enzyme
storage stability. It is possible that aggregation within the viscosity modifiers or diffusion
of H2O2 might have played a role in keeping the enzyme active for longer times. With the
exception of the results corresponding to the fridge storage, in general it is observed that
through physical immobilization on paper, the HRP enzyme is capable of retaining its
biological activity intact for significantly longer periods of time compared to solution (40
days, see Figure 4.5).
4.4 – CONCLUSIONS
To formulate an enzyme-containing bio-ink for piezoelectric inkjet printing,
surface tension and viscosity should be matched to the requirements of the
printhead design without undermining post-printing functional objectives.
Viscosity modifiers typically used in commercial ink formulations were found to
significantly affect the initial activity of the HRP-containing bio-ink, possibly
due to diffusion limitations. CMC, a charged polymer, appears to be an effective
viscosity modifier that is capable of increasing the ink viscosity to the desired
values for piezoelectric inkjetting without deteriorating the biological activity of
the enzyme. The ink additives did not significantly affect the storage stability of
the bio-ink.
Horseradish peroxidase can be reliably deposited on solid supports using a
piezoelectric inkjet printer if ink formulation and jetting parameters are
optimized. An operational window for reliable and uniform drop formation was
found for the bio-ink formulation containing 0.5 wt-% of CMC as the viscosity
modifier; however, printing speed is considerably lower than what is reported
for typical commercial inkjet printers for graphic arts applications. First drop
problems were avoided by adding a humectant to the ink formulation. The high
shear rate during jetting did not significantly affect the enzyme activity.
Uncoated wood-free paper inkjet printed with HRP enzyme can be used in H2O2
detection without a loss of performance for 3 months when stored at 23°C and
50%RH and for at least 8 months when stored in a freezer.
— 96 —
Chapter 5
FIBROUS MATERIALS AS SUPPORTS FOR BIOACTIVE PAPERS
5.1-INTRODUCTION
The solid support where biomolecules are immobilized plays a significant role in the
performance of bioanalytical systems [27]. Studies on the effects of immobilization on
enzymatic reactions have suggested that properties of the immobilization support strongly
affect enzyme activity and stability over time [212]. When restricted in spatial movement,
proteins can adopt new configurations and the degree of accessibility to the active site
can change. Moreover, the solid support can influence the spatial distribution and the
binding of the bioagent can affect, in turn, post print detection and bioanalytical
performance.
Among the wide range of supports available for enzyme immobilization,
cellulosic fibrous materials are attractive because they offer: a porous structure
with a large surface area for larger biomolecule loads, a suitable environment for
biorecognition events, low cost, portability, and disposability. As reviewed in
section 2.2.3, cellulose and its derivatives have been extensively used as
supports for biomolecule immobilization [32,36], whereas paper has been
incorporated in dry reagent chemistries either as a preformed matrix or as a
reflective layer [37-42] and as a support for paper-based microfluidic patterning
in bioanalysis [40-44]. However, no systematic study of the effect of cellulosic
paper support characteristics on HRP enzyme activity and bioanalytical
performance was found in the literature.
5.2-SPECIFIC OBJECTIVE
The goal of this chapter is to systematically study the performance of paper as an enzyme
immobilization support in bioanalysis. To achieve this objective, the HRP bio-ink
formulated in the Chapter 3 was inkjet printed using the jetting conditions investigated in
— 97 —
the same chapter and was physically immobilized on various fibrous substrates. The
impact of the paper supports on the performance of hydrogen peroxide bioanalysis was
investigated and the localized enzyme activity within the printed papers was
characterized. As spreading and penetration of the bio-ink appeared to be a significant
factor in the bioanalytical response of the studied papers, paper handsheets with an
increasing hydrophobicity level obtained by addition of increasing doses of the sizing
agent were also studied.
A version of this chapter has been accepted for publication in the Journal of Pulp and
Paper Science (November 2008).
5.3 - RESULTS
5.3.1 - Bioanalysis of H2O2 on fibrous substrates
5.3.1.1 - Color profile
The standard HRP bio-ink formulation (see section 3.1.6) was printed on the
commercial papers detailed in Table 3.2 using the Dimatix DMP 2800 printer.
Square spots were patterned with 1200 d.p.i. spatial resolution using the jetting
conditions described in Table 4.2. Taking into account the concentration of
enzyme in the bio-ink, the jetted drop volume, and the print resolution, the HRP
deposited amount per unit area of support was 118 nmol/m2.
The printed papers were used to perform solid phase bioanalysis of a 2 mM
H2O2 solution following the procedure described in section 3.5.2. Table 5.1
illustrates the typical color spots generated on the different commercial papers
and the wavelength corresponding to the maximum reflectance from the diffuse
reflectance spectra (See Appendix A.7). In Table 5.1, the results are presented
for short (<1 h) and for long (> 1 h) time periods after exposure to the analyte.
— 98 —
Table 5.1 – Color response developed by different papers printed with standard
bio-ink after exposure to a 2 mM H2O2 solution. Paper codes: (A)
chromatographic paper, (B) uncoated mechanical paper, (C) uncoated recycled
paper, (D) uncoated wood-free paper, (E) color copy cover, (F) coated grade for
offset, (G) coated grade for inkjet, and (H) cast coated paper.
Paper Code
Peak Reflectance Wavelength [nm]
Scanned Color Response
<1h >1h Blank <1h >1h
A 480 480
B 700 700
C 480 700
D 480 480
E 480 480
F 480 480
G 480 700
H 700 700
— 99 —
Qualitative differences amongst paper supports become apparent from the
comparison. If color is measured after 20 min (to guarantee invariable color
intensity) and within 1 h of exposure to the analyte, most of the papers exhibit a
reflectance peak at 480 nm that is characteristic of the bluish shade of green
color developed by the oxidized ABTS specie. In contrast, the cast coated paper
(H) and the paper containing mechanical pulp (B), immediately develop a pink
color with a peak reflectance at 700 nm. These results suggest that for similar
conditions of exposure (time, analyte concentration) to H2O2, papers differed in
their color profiles (wavelength corresponding to maximum reflectance). In
addition, the color profile developed by some of the papers (C and G) was time-
dependent (peak reflectance wavelength changed after 1h) going from bluish
green to pink. Moreover, it was observed that the color intensity (magnitude of
the peak reflectance) for the same analyte concentration, time of exposure, and
peak wavelength was paper-dependent.
No clear explanation for the different paper color profiles is available. Some possibilities
that require further explorations are: a) presence in the paper of reducing agents
competing with ABTS for the analyte, b) presence in the paper of oxidizing agents other
than H2O2 that further extend the ABTS reaction or, c) generation of pink HRP
intermediate compounds generated under conditions of substrate inhibition [24].
5.3.1.2 - Bioanalytical performance
Reflectance spectrophotometry was used to quantify the color developed by the
printed areas upon exposure to H2O2 solutions with increasing concentrations,
following the procedure detailed in sections 3.5.3 and 3.5.4.
Figure 5.1 shows a typical calibration plot of AR vs. LogC corresponding to
peak diffuse reflectance (480nm) of the color spots developed by paper D when
exposed to increasing analyte concentrations. Calibration plots were not
obtained for papers B (uncoated mechanical paper) and H (cast coated paper)
because the initial color profile was not comparable to the rest of the papers and
neither for paper A (chromatographic paper) because ink spreading made the
— 100 —
color intensity highly variable. A similar type of calibration plot was obtained
for the remaining papers (See Appendix A.8).
Figure 5.1- Calibration curve of AR vs. log C for H2O2 bioanalysis performed on
uncoated wood free paper printed with HRP bio-ink. C = log(109.[H2O2]) mol.l-1. Error
bars represent the ±one standard deviation in the AR value corresponding to the
simultaneous color development of ten enzyme spots printed on the same paper strip.
Figure 5.1 includes performance indicators for quantitative bioanalysis: the
lower limit of detection defined as the concentration of analyte corresponding to
the intercept of the calibration curve with the abscissa axis; the upper limit of
detection defined as the maximum concentration of analyte that allowed
maximum linear correlation between AR and log C in the calibration curve; the
linear range of measurement obtained from the difference between upper and
lower limits of detection; and the sensitivity of the technique represented by the
slope of the calibration curve.
Figures 5.2 and 5.3 summarize the indicators of bioanalytical performance for
all the papers studied.
y = 0.43x - 2.49
R2 = 0.99
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
1 2 3 4 5 6 7 8 9 10 11 12
Log C
AR
Stoichiometry
36.4mM
Lower Limit of
Detection
Upper Limit of
Detection
Sensitivity
— 101 —
Figure 5.2 - Sensitivity for H2O2 solid phase bioanalysis on different commercial paper
supports. Paper codes: (C) uncoated recycled paper, (D) uncoated wood-free paper, (E)
color copy cover, (F) coated grade for offset, and (G) coated grade for inkjet.
Except for paper E (color copy cover), in general, the commercial papers did not
differ much in their lowest limit of detection. However, the sensitivity of the
method performed on different papers varied significantly. The uncoated wood
free paper (D) was the most sensitive, meaning that for a similar analyte
concentration, this paper developed the highest color intensity in the series. The
second highest sensitivity was shown by the coated paper for offset (F). The
largest linear range of measurement pertained also to the uncoated wood-free
paper.
0
0.1
0.2
0.3
0.4
0.5
C D E F G
Samples
Se
nsit
ivit
y
— 102 —
Figure 5.3 - Limits of detection and linear range of measurement for H2O2 solid
phase bioanalysis on different commercial paper supports. Open and closed
squares correspond to lower and upper limits of detection, respectively. Paper
codes: (C) uncoated recycled paper, (D) uncoated wood-free paper, (E) color
copy cover, (F) coated grade for offset, and (G) coated grade for inkjet.
Interestingly, paper D (uncoated wood-free paper) was the more hydrophobic
paper in the subgroup of the uncoated papers (as measured by HST method,
results shown in Appendix A.9), and paper F (coated paper for offset) was the
paper containing the most hydrophobic binder (latex as opposed to polyvinyl
alcohol) in the coated subgroup of papers. This observation helped narrow down
the grade of papers that justified further exploration and suggested that
hydrophobicity could have played a role in the bioanalytical performance.
Moreover, differences in the paper structure could have produced variations in
the local enzyme concentration amongst the papers causing distinctive
quantitative color responses.
For H2O2 bioanalysis, the degree of correlation between the color intensity
signal and the analyte concentration was paper-dependent. The wood free
uncoated paper (D) showed the best degree of correlation with R²=0.99. This
5
6
7
8
C D E F G
Samples
Lim
its
of
De
tec
tio
n
(L
og
C)
— 103 —
value is similar to values reported in the literature for chemical analysis of
inorganic cations on a filter paper [150].
5.3.2 - HRP cross-sectional distribution in commercial papers
A new HRP bio-ink formulation (see section 3.6.2) containing a fluorescent
probe was printed on the commercial papers detailed in Table 3.2 using the
Dimatix DMP 2800 printer. Square spots with 1200 d.p.i. spatial resolution were
printed using the jetting conditions described in Table 4.2. Taking into account
the concentration of enzyme in the bio-ink, the jetted drop volume and the print
resolution, the HRP deposited amount per unit area of support was 118 nmol/m2.
The printed papers were exposed to a 0.1 mM H2O2 solution to develop the
fluorescent response and then embedded, cross-sectioned, and imaged using
CLSM as detailed in section 3.6.
The cross-sectional distribution of active HRP was examined for samples C to
G; (C) uncoated recycled paper, (D) uncoated wood-free paper, (E) color copy
cover, (F) coated grade for offset, and (G) coated grade for inkjet). In the next
page, Figure 5.4 shows the combined paper filler/pigment-enzyme distribution
for a negative control, an inkjet printed sample, and a positive control of the
papers. High refractive index species (pigment or fillers) are shown in green and
active HRP enzyme is shown in red.
— 104 —
Figure 5.4 - Combined cross-sectional CLSM images of the active HRP enzyme (red) and
pigments/fillers (green) for negative controls, inkjet printed sample, and positive control of
commercial papers. Paper codes: (C) uncoated recycled paper, (D) uncoated wood-free paper,
(E) color copy cover, (F) coated grade for offset, and (G) coated grade for inkjet. For paper E
only partial cross-sectional view is shown. Bar = 50µm.
— 105 —
First, Figure 5.4 shows that some of the printed samples (uncoated wood-free
paper, D; coated paper for offset, F; and G, coated paper for inkjet) exhibited
partial penetration of the bio-ink in the thickness direction; whereas the other
samples (uncoated recycled paper, C and color copy cover, E) are fully
penetrated by the bio-ink. It follows that the local concentration of enzyme in
the papers with partial thickness penetration of the bio-ink is higher than in the
papers with full penetration. This observation is consistent with the enhanced
color response detected in papers with a higher degree of sizing.
Second, within the CLSM lateral resolution (116nm/pixel), it can be observed
that for all the paper samples, the enzyme appeared to be preferentially located
in the fibers and not in the inorganic pigment or filler domains (red and green
areas do not overlap). The result is particularly evident for the two-sided inkjet
grade coated paper (paper G): the coating layers and the filler do not show
detectable enzyme activity. It appears that cellulosic fibers present a more
suitable environment for the enzymes than the inorganic pigments and fillers.
The cell wall of the swollen delignified cellulose fibers is characterized by rather
monodisperse microvoids with an average diameter around 100nm [213]. The
microvoids in the cellulose fiber wall exist while the fiber is wet and disappear
as the structure dries and shrinks. The HRP molecule, on the other hand, is an
ellipsoid with 6.5 nm x 5.4 nm x 4.3 nm main dimensions [21] that can move
with water as long as it does not become physically entrapped or chemically
bound. The microvoids in the cellulose fiber are the smaller pores available in
the paper structure to entrap the enzyme and because they are dynamic pores,
they can potentially lock the enzyme inside the fiber cell wall as the paper dries.
In addition, cellulose has hydroxyl and carboxyl groups that can interact with
some of the amino acids in the enzyme surface, increasing the enzyme-cellulose
affinity; this point will be explored further in Chapter 6 of the thesis. These
cellulose fiber attributes may explain the higher concentration of enzyme found
in the fibers.
— 106 —
5.3.3 - HRP cross-sectional distribution in coating layers
To further investigate differences in the enzyme distribution between inorganic
pigments/fillers and fibers, a thick layer of coating (10pph SB latex and 100pph
ground calcium carbonate) was prepared following the steps detailed in section
3.2.1.3. The coating layer was then printed and examined under the same
experimental conditions as the commercial paper samples.
Figure 5.5 shows the pigment (green) and enzyme (red) distributions and their
combined visualization. Interestingly, the enzyme moves in the same direction
as the analyte diffuses, away from the bulk of the coating mainly remaining on
the surface of the coating layer. It seems that the pores in the coating structure
are too big to entrap the enzyme and it can freely flow through the coating layer
without significant binding. Moreover, it might be speculated (is not possible to
make direct observation from CLSM images due to lack of enough spatial
resolution) that there is low affinity between the enzyme and the coating layer.
Further insight into this point will be provided in Chapter 6.
Figure 5.5 - CLSM cross-sectional images of a thick coating layer with HRP
enzyme printed on the left side and exposed to H2O2 solution on the right side.
Left: pigment map. Center: active HRP enzyme map. Right: overlay of pigment
and enzyme maps.
— 107 —
5.3.4 - HRP cross sectional distribution in handsheets with an increasing degree of
internal sizing
A pattern of dots with a 100 d.p.i. spatial resolution was printed on paper
handsheets with an increasing level of hydrophobicity using the Amplex Red-
containing bio-ink and the jetting conditions described in Table 4.2. After
fluorescence development with a 0.1 mM H2O2 solution the printed surface was
imaged with CLSM according to the procedure detailed in section 3.6.5.
In the next page, Figure 5.6 shows the surface view of an active HRP enzyme
distribution obtained using CLSM for a series of handsheets prepared with doses
between 0 and 1.6wt-% of a rosin-based sizing agent (mixture of resin acids
mainly composed of abietic acid, see HST degree of sizing in Appendix A.10).
For sizing levels lower than 0.8wt-%, the bio-ink spreads all over the handsheet
and no detectable red fluorescence is observed due to the very low local enzyme
concentration. As the sizing level increases from 0.8wt-% to 1.6wt-%, for each
drop of bio-ink delivered by the printer, a red fluorescent printed dot is detected
after exposure to the analyte.
— 108 —
Figure 5.6 - Combined surface view CLSM images of the active HRP enzyme
(red) and fibers (green) for handsheets increasingly sized from 0 to 1.6wt-% of a
rosin based sizing agent. Bar= 400µm.
Moreover, Figure 5.7 illustrates the bio-ink cross-sectional distribution for the
handsheets that exhibited well-defined red-fluorescent printed dots. As expected,
the CLSM images show a decrease in the cross-sectional bio-ink penetration as
the sizing level increases. However, from Figure 5.6, it also appears that
oversizing the fibers (e.g., 1.2wt-% and 1.6wt-%) does not further reduce the
degree of spreading or intensifies the red fluorescent response. It appears that
although minimized spreading and penetration should maximize the local
enzyme concentration, the highly hydrophobic fibrous support might have
partially inactivated the HRP enzyme. The impact of support hydrophobicity on
enzyme biological functionality is investigated further in Chapter 6.
— 109 —
Figure 5.7 - Combined cross-sectional CLSM images of the active HRP enzyme
(red) and fibers (green) for handsheets increasingly sized from 0.8 to 1.6wt-%.
Bar = 50µm.
Similarly, the spatial distribution of the active HRP on printed handsheets sized
with up to 0.3wt-% of an AKD-based sizing agent was examined (See HST
degree of sizing in Appendix A.10). Figure 5.8 illustrates the CLSM images
corresponding to the surface view of the inkjet printed samples. Red fluorescent
dots are detected above 0.05wt-% AKD. Qualitatively, the fluorescence intensity
of the printed spots corresponding to well sized fibers is lower for AKD-sized
handsheets (0.3wt-%) when compared to rosin-sized handsheets (0.8wt-%)
suggesting that the type of hydrophobizing agent may play a role in affecting the
activity of the enzyme.
Figure 5.8 - Combined surface view CLSM images of the active HRP enzyme
(red) and fibers (green) for handsheets increasingly sized from 0 to 0.3wt-%
with AKD based sizing agent. Bar= 400µm.
0% 0.05% 0.3%
— 110 —
5.4 – CONCLUSIONS
Papers differ qualitatively and quantitatively as solid supports for the solid phase
bioanalysis of H2O2 using the enzyme HRP and the chromogenic co-substrate
ABTS. This observation indicates that the detection function of bioactive papers
is significantly affected by the support properties. The differences in
performance can be explained by differences in both structure and surface
chemistry of the papers.
A new technique based on CLSM allowed the characterization of the active
enzyme spatial distributions in naturally fluorescent paper substrates. CLSM
images suggest that partial penetration of the bio-ink and minimum spreading
favor the bioanalytical response. Also, HRP enzyme preferentially locates in the
fiber cell wall and not near the pigments or fillers. It appears that the microvoids
in the fiber cell wall may help entrap the enzyme and present a more suitable
microenvironment for the preservation of enzyme biological functionality.
The impact of surface chemistry was studied by increasing the sizing level of a
series of handsheets. It was confirmed that spreading and penetration can be
controlled to enhance the colour response by maximizing the local active
enzyme concentration. However, a limit exists to the amount of sizing agent that
can be added. Oversizing the fibers can lead to a partial inactivation of the
enzyme and, hence, a reduced bioanalytical performance. In addition, the local
enzyme activity on well sized fibers seems to depend on the type of sizing agent.
These effects will be explored in more depth in Chapter 6.
The uncoated wood free paper (paper D) presented the best overall bioanalytical
performance due to its highest sensitivity and widest linear range of detection in
comparison with the rest of the commercial papers. Coated papers with their
improved surface for printing capable of retaining most of the ink on the surface
are considered high quality supports for graphic arts applications. In contrast,
pigment coatings seem not to contribute favorably to the analytical performance
of bioactive papers.
— 111 —
Chapter 6
PAPER-ENZYME INTERACTIONS
6.1-INTRODUCTION
The interactions between the biomolecules and the cellulosic fiber surfaces
together with the associated interfacial phenomena are critically important in
developing bioactive papers because they govern the distribution, binding
behavior, biological activity, and stability of the immobilized biomolecule, and
therefore, the bioanalytical performance of the final product.
Adsorption is the simplest immobilization strategy for attaching biomolecules,
such as enzymes, on to solid supports. However, because paper is a porous and
chemically heterogeneous support, it is difficult to predict the adsorption
behaviour of the biomolecules. Even for the case of pure physical adsorption,
the existence of both external and internal surfaces within the cellulosic fibers
renders the interactions between the biomolecules and the substrate highly
complex.
In the literature, adsorption of proteins on surfaces has been extensively
investigated [139-143]. But, with the exception of cellulases, little is known
about the interactions between proteins, in particular enzymes, and cellulosic
fiber surfaces, as well as with other key components and additives used in
papermaking. Investigating the adsorption behavior of enzymes on relatively
simple model surfaces can provide some insights into the nature and extent of
the interfacial interactions that can yield both larger adsorbed enzyme amounts
and a higher immobilized enzyme activity. As a result, solid paper supports with
the most suitable surface chemistry for bioactive paper applications can be
engineered.
— 112 —
6.2-SPECIFIC OBJECTIVE
The goal of this thesis chapter was to characterize the adsorption behavior and activity of
the enzyme HRP on cellulosic fiber surfaces to better understand the role that
hydrophobic and electrostatic interactions play in enzyme immobilization. A better
understanding of the fundamental interactions between the enzyme and the cellulosic
fiber surfaces can provide insights into the cause for differences in performance of the
various paper supports studied and the local active enzyme distributions observed in
Chapter 5. To achieve this objective, HRP adsorption on model sorbents with different
surface energies, surface charge signs, and surface charge densities was examined. HRP
adsorption on cellulosic fibers internally sized with different types and doses of
hydrophobizing agents was also investigated.
6.3 – RESULTS
6.3.1 - HRP adsorption and activity on model sorbents
Adsorption experiments were conducted using HRP solutions with concentrations
between 0 and 0.4 mg/ml and model sorbents (see section 3.2.2.1) selected to have the
following features:
• Low or no porosity to favor immobilization by physical adsorption over entrapment
• Relative uniformity in surface chemistry
• High specific surface area to reduce bioagent volumes (cost)
• Range of surface free energies
Table 6.1 summarizes properties of the model sorbents used in this investigation and
Figure 6.1 illustrates the morphology of the sorbents through SEM micrographs.
— 113 —
Table 6.1 - Properties of the model sorbents
Sorbent γD [mN/m]a γ
P [mN/m]b γT [mN/m]c ζ [mV]d SSA [m2/g]e
Polystyrene 38.4f 4.2f 42.6f -59.7±6.1 14.70
Microcrystalline cellulose 52.94g 4.24g 57.18g -4.3±2.3 1.76
SB Latex 38.2h 16.9h 55.1h -52.2±2.4 0.633
Regenerated cellulose (rayon) 23.75i 26.31i 50.06i -13.8±3.6 19.48
Calcium carbonate 35.2j 32.5j 67.7j -33.3±2.3 12.50
Notes: a dispersive component of surface free energy extracted from literature
b polar component of surface free energy extracted from literature
c total surface free energy extracted from literature d measured zeta potential in 40mM KH2PO4 buffer pH 6.8 using method described in section 3.2.2.3.3 corresponding to the average ± one standard
deviation of three independent measurements e specific surface area f [166] g [171] h [178] i [174] j [176]
— 114 —
Figure 6.1. SEM micrographs of model sorbents a) Microcrystalline cellulose (400X), b) Ground rayon filament yarn (200X), c)
Ground calcium carbonate (4500X), d) Polystyrene beads (20000X), and e) SB latex (3000X).
— 115 —
The depletion method described in section 3.7.2 was used to characterize the adsorbed
amount of protein and the residual enzyme activity of HRP after adsorption. Figures 6.2
and 6.3 show the HRP adsorption and inactivation isotherms on the model sorbents
constructed according to the procedures outlined in sections 3.7.3 and 3.7.4,
correspondingly.
Figure 6.2. Adsorption isotherms of HRP on model surfaces. Error bars represent the ±
one standard deviation for two independent adsorption experiments. Protein
measurements for each adsorption experiment represent an average of two sample
measurements.
Figure 6.2 shows that HRP adsorption behavior is considerably affected by the nature of
the sorbent. The more hydrophobic sorbents (polystyrene and microcrystalline cellulose)
exhibit a high adsorption affinity (higher initial slope in the adsorption isotherm) and a
larger amount of protein adsorbed, indicating a stronger enzyme-sorbent interaction,
whereas HRP binds weakly to the more hydrophilic sorbents (rayon and ground calcium
carbonate) and the surfaces load HRP in very low amounts. Moreover, increased enzyme
-0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0 0.1 0.2 0.3 0.4 0.5
HR
P ad
sorb
ed a
mou
nt [m
g/m
²]
HRP equilibrium concentration [mg/ml]
Polystyrene
Microcrystalline cellulose
SB Latex
Rayon
Ground Calcium Carbonate
— 116 —
adsorption (Figure 6.2) consistently corresponds to a decreased sorbent polar surface free
energy (Table 6.1). These results are in agreement with the general observation that
proteins are adsorbed in larger amounts on hydrophobic surfaces [144].
In particular, the HRP adsorption behavior on calcium carbonate and SB latex helps
explain the HRP spatial distribution and bioanalytical performance observed in pigment
coated papers (Chapter 5). Pigment coated papers are mainly composed of inorganic
pigments, such as calcium carbonate, and a small amount of binder, such as SB latex, to
hold the pigment particles together. It is clear from Table 6.1 that calcium carbonate is
the most hydrophilic sorbent in the series and has the largest polar surface free energy
component. Also, from Figure 6.2 it is observed that HRP-calcium carbonate affinity is
very low (scant bioagent uptake by the sorbent), resulting in a low local enzyme activity
in coating layers, and therefore, poor performance during bioanalysis. Notwithstanding, it
is plausible to expect that the more hydrophobic coating component, SB latex, will help
bind the enzyme more strongly. However, just a few points of the HRP adsorption
isotherm on SB latex (see Figure 6.2) suggest that although SB latex shows a higher
enzyme load than GCC, the adsorption isotherm is not of the high affinity type. This
observation added to the fact that pigmented coatings incorporate very low amounts of
latex explain the poor behavior of coated papers as HRP sorbents. These results suggest
that pigment coated papers will probably be less suitable enzyme supports in bioactive
paper applications. In contrast, uncoated papers treated to render them more hydrophobic
hold promises for bioactive paper development.
As explained in section 3.4.1, ±10% experimental error is typically found in HRP activity
tests in solution. Activity tests corresponding to independent adsorption experiments tend
to show an increased variability (±20% error) due to the complexity of the sorbents. In
this context, Figure 6.3 shows no significant lose of enzyme activity upon adsorption on
most of the sorbents within the experimental error. However, polystyrene, the most
hydrophobic support in the series, produces a considerable inactivation; which is most
pronounced at low enzyme concentrations. As expected, the larger the available surface
area in relation to the availability of enzyme in solution is, the higher the surface-induced
changes in the enzyme. These observations are in agreement with previous research
— 117 —
indicating that stronger hydrophobic interactions can cause extensive conformational
changes in enzymes, leading to inactivation and denaturation [145, 214,215]. Higher HRP
adsorbed amounts per unit area tend to minimize the surface-induced inactivation on
polystyrene suggesting that the enzyme molecules when surrounded by other enzyme
molecules tend to preserve their native folded state. A better understanding of the
interplay between activity and adsorption behavior can help define the desirable
conditions for bioactive paper applications (high adsorption, low inactivation).
Figure 6.3 - Inactivation isotherms of HRP on model surfaces. The ±20% area is the
uncertainty range for the inactivation measurements. Error bars represent the ± one
standard deviation for two independent adsorption experiments. Protein and activity
measurements for each adsorption experiment represent an average of two sample
measurements.
6.3.2 - Impact of the type of surface charge on HRP adsorption and activity
Rayon, negatively charged non-porous amorphous cellulose, was selected as the sorbent
to evaluate the impact of electrostatic interactions on HRP adsorption. The untreated
-40
-20
0
20
40
60
80
100
120
0 0.1 0.2 0.3 0.4 0.5
HR
P in
acti
vati
on [%
]
HRP equilibrium concentration [mg/ml]
Microcrystalline cellulose
Rayon
Ground Calcium Carbonate
Polystyrene
SB Latex
— 118 —
sorbent had a zeta potential of -39.1±2.4 mV (corresponding to the average ± one
standard deviation, three independent measurements). After surface charge modification
by layer deposition of the polyelectrolyte PAH (see section 3.2.2.2.4), rayon with a
positively charged surface of 18.4±1.4 mV (corresponding to the average ± one standard
deviation, three independent measurements) was obtained.
As in the previous experiments, the depletion method was used to characterize the
adsorbed amount of protein and the residual enzyme activity of HRP after adsorption on
the two oppositely charged sorbents. HRP adsorption and inactivation isotherms to the
model sorbents are shown in figures 6.4 and 6.5, respectively.
Figure 6.4 - Adsorption isotherms of HRP on rayon with positive and negative surface
charges. Error bars represent the ± one standard deviation for two independent adsorption
experiments. Protein measurements for each adsorption experiment represent an average
of two sample measurements.
If the HRP adsorption isotherms on rayon and microcrystalline cellulose from Figure 6.2
are compared against the HRP adsorption isotherms on positive and negative rayon from
0.0
0.1
0.2
0.3
0.0 0.1 0.2 0.3 0.4 0.5
HR
P ad
sorb
ed a
mou
nt [m
g/m
²]
HRP equilibrium concentration [mg/ml]
Negative rayon
Positive rayon
— 119 —
Figure 6.4, it can be concluded that electrostatic interactions affect to a much lesser
extent the loading and adsorption affinity of HRP than hydrophobic interactions.
Moreover, positively charging the rayon surface does not seem to contribute favorably to
HRP adsorption. This outcome might be justified by results obtained by isoelectric
focusing and electrophoresis showing a main band of peroxidase activity at pH 8.5 (two
more bands at pH 3.5 and 6.8) in the commercial HRP enzyme (Sigma Type VI) [216].
Under the adsorption experimental conditions (pH=6.8), the enzyme has a net positive
surface charge; hence, it adsorbs more onto the negative rayon. This result also provides
an interesting observation for papermakers since cationic polyelectrolytes are routinely
used in papermaking processes as a retention aid for keeping the fillers in the paper.
Figure 6.5 - Inactivation isotherms of HRP on rayon with positive and negative surface
charges. The ±20% area is the uncertainty range for the inactivation measurements.
Error bars represent the ± one standard deviation for two independent adsorption
experiments. Protein and activity measurements for each adsorption experiment represent
an average of two sample measurements.
-50
-40
-30
-20
-10
0
10
0.0 0.1 0.2 0.3 0.4 0.5
HR
P in
acti
vati
on [%
]
HRP equilibrium concentration [mg/ml]
Negative rayon
Positive Rayon
— 120 —
From Figure 6.5, no significant loss of enzyme activity upon adsorption is observed,
within the experimental error (±20%). Interestingly, for the positive rayon sorbent some
observations with an increased activity after adsorption are found. Also, a large
variability is seen in these observations that might be explained by an incomplete
coverage of the fiber surface with the positive polyelectrolyte.
6.3.3 - Impact of the surface charge density on HRP adsorption and activity
Rayon, a negatively charged non-porous amorphous cellulose, and BKSW, a negatively
charged porous cellulosic fiber, were selected as sorbents to evaluate the impact of
increases in negative surface charge density (electrostatic interactions) on HRP
adsorption. After TEMPO-mediated oxidation of the rayon and the BKSW fibers, an
increment in the negative surface charge density through the increase in the carboxyl
content was obtained (See section 3.2.2.2.3). Table 6.2 summarizes the properties of the
sorbents.
As in the previous experiments, the depletion method was used to characterize the
adsorbed amount of protein and the residual enzyme activity of HRP after adsorption on
the untreated and oxidized sorbents. HRP adsorption and inactivation isotherms to the
model sorbents are shown in Figures 6.6 and 6.7, respectively.
Table 6.2 - Properties of the rayon and fibers with and without oxidation treatment
Sorbent Carboxyl content [meq/g] SSA [m2/g]
Rayon 0.05 19.48
Oxidized rayon 0.37 778.85
BKSW fibers 0.08 22.27
Oxidized BKSW fibers 0.29 649.54
— 121 —
Figure 6.6 - Adsorption isotherms of HRP on rayon and cellulosic fibers with and
without TEMPO-mediated oxidation treatment. Error bars represent the ± one standard
deviation for two independent adsorption experiments. Protein measurements for each
adsorption experiment represent an average of two sample measurements.
From Figure 6.6, it can be concluded that irrespective of the sorbent (rayon or pulp fiber),
an increase in the negative surface charge density in the sorbent results in a decrease in
the amount of enzyme loading. If the HRP adsorption isotherms on rayon and
microcrystalline cellulose from Figure 6.2 are compared against the HRP adsorption
isotherms on untreated and oxidized rayon from Figure 6.6, it is clear that electrostatic
interactions affect to a much lesser extent the loading and adsorption affinity of HRP than
hydrophobic interactions. Moreover, the increment in the surface carboxyl content in the
sorbent produces a considerable augmentation in the specific surface area, which
enhances the HRP loading capacity per unit mass of sorbent. However, the HRP loading
capacity per unit area decreases, indicating that a larger negative surface charge density
does not seem to contribute favorably to HRP adsorption affinity. This outcome might be
justified by the fact that this particular type of commercial HRP C enzyme has been
0.00
0.05
0.10
0.15
0.0 0.1 0.2 0.3 0.4
HR
P a
dsor
bed
amou
nt [
mg/
m²]
HRP equilibrium concentration [mg/ml]
Untreated rayon
Oxidized rayon
Untreated cellulosic fibers
Oxidized cellulosic fibers
— 122 —
chemically modified to protect their amino (positive) groups. Therefore, the positive
amino groups that can electrostatically interact with the negative carboxyl groups of the
sorbents are not available.
Interestingly, there seems to be a shift in the isotherm curvature from almost linear
(untreated rayon and pulp fibers) to concave (oxidized rayon and pulp fiber). This change
in curvature could be associated with a change in the binding mechanism, the presence of
more than one adsorbed layer, a change in protein orientation, an increased binding
anisotropy, or a competitive adsorption between isoenzymes.
Figure 6.7 - Inactivation isotherms of HRP on rayon and cellulosic fibers with and
without TEMPO-mediated oxidation treatment. The ±20% area is the uncertainty range
for the inactivation measurements. Error bars represent the ± one standard deviation for
two independent adsorption experiments. Protein and activity measurements for each
adsorption experiment represent an average of two sample measurements.
From Figure 6.7, no significant loss of enzyme activity upon adsorption is observed
within the experimental error (±20% error). Interestingly, some observations of an
-70
-60
-50
-40
-30
-20
-10
0
10
20
30
40
0.0 0.1 0.2 0.3 0.4
HR
P in
acti
vati
on [
%]
HRP equilibrium concentration [mg/ml]
Untreated rayon
Oxidized rayon
Untreated cellulosic fibers
Oxidized cellulosic fibers
— 123 —
increased activity after adsorption are found. Also, a large variability is seen in the
experimental observations corresponding to the cellulosic fibers and that might be
explained by the higher structural and chemical complexity of these sorbents compared to
rayon. In addition, lower protein adsorbed amounts are expected to be affected more by
the experimental error.
6.3.4 - Impact of internal sizing on HRP adsorption and activity
During papermaking, internal sizing agents are typically added to diluted suspensions of
fibers in water to decrease their surface energy, mainly the polar component [217, 218],
and to reduce water absorption and swelling [219]. Since hydrophobic interactions were
previously found to result in larger HRP loadings onto the model sorbents, it was relevant
to investigate the adsorption behavior of HRP on cellulosic fibers with an increasing
degree of hydrophobicity. Table 6.3 summarizes the properties of the fibers before and
after the sizing treatment.
As before, the depletion method was used to characterize the adsorbed amount of protein
and the residual enzyme activity of HRP on fibers hydrophobized (internally sized) with
a rosin-based (mixture of resin acids mainly composed of abietic acid) sizing agent and
an AKD sizing agent (see section 3.2.2.2). It is important to note that pulp fiber as a
sorbent material shows more complexity in their structure than the model sorbents due to
both the presence of significant porosity (lumen and microvoids in the cellulosic fiber
cell wall) and a more heterogeneous chemistry (amorphous and crystalline cellulose).
Thus, enzyme immobilization proceeds by adsorption on external and internal surfaces,
in regions with organized and disorganized cellulose chains.
HRP adsorption and inactivation isotherms on the untreated and treated pulp fibers are
shown in Figures 6.8 and 6.9, respectively.
— 124 —
Table 6.3 – Properties of the fibers
Fiber treatment γD [mN/m]a γ
P [mN/m]b γT [mN/m]c HST [s]d ζ [mV]e SSA [m2/g]f
Untreated 34.05 6.456 40.51 1 -18.6 ± 5.0 22.27
0.8 wt-% rosin 28.4 0.214 28.4 452 -15.7 ± 3.3 22.27
1.6 wt-% rosin 27.65 0.178 27.65 2388 -20.1 ± 1.0 22.27
0.3 wt-% AKD 25.25 0.39 25.25 5000 -12.1 ± 1.1 22.27
Notes: a dispersive component of surface free energy measured by dynamic contact angle
b polar component of surface free energy measured by dynamic contact angle c total surface free energy measured by dynamic contact angle d degree of sizing measured with Hercules Sizing Test, 80% reflectance, 1% formic acid e measured zeta potential in 40mM KH2PO4 buffer pH 6.8 using method described in section 3.2.2.3.3 corresponding to the average ± one standard deviation of three independent measurements f specific surface area measured by methylene blue method
— 125 —
Figure 6.8 - Adsorption isotherms of HRP on beaten bleached kraft softwood fibers with
an increasing degree of internal sizing (hydrophobicity). Error bars represent the ± one
standard deviation for two independent adsorption experiments. Protein measurements
for each adsorption experiment represent an average of two sample measurements.
As found with the model sorbents, an increase in the protein loading is found for the
more hydrophobic fibers (treated with the rosin-based sizing agent, see Figure 6.8) when
compared to the untreated fibers. The adsorption affinity also increases, but in contrast to
polystyrene (Figure 6.2), the adsorption isotherms remain of the low affinity type. Again
increased adsorption consistently corresponds to a lower sorbent polar component of the
surface free energy. The AKD-treated fibers do not show a considerable improvement in
adsorbed amounts or adsorption affinity with respect to the control fibers. This
observation agrees with the qualitatively low fluorescence response found in AKD-sized
handsheets and reported in Chapter 5. Although AKD-sized fibers are highly
hydrophobic in terms of degree of sizing (HST method), they exhibit a larger polar
surface free energy component when compared with the rosin-sized fibers.
0.0
0.1
0.2
0.3
0.4
0.0 0.1 0.2 0.3 0.4 0.5
HR
P a
dsor
bed
amou
nt [
mg/
m²]
HRP equilibrium concentration [mg/ml]
Untreated
0.8 wt-% rosin-based sizing
1.6 wt-% rosin-based sizing
0.3 wt-% AKD sizing
— 126 —
Figure 6.9 illustrates the impact of the different fiber treatments on the residual activity of
the enzyme. Except for some isolated observations showing enzyme activation upon
adsorption, no statistical differences in activity can be inferred from the data.
Figure 6.9 - Inactivation isotherms of HRP on beaten bleached kraft softwood fibers
model surfaces with an increasing degree of internal sizing (hydrophobicity). The ±20%
area is the uncertainty range for the inactivation measurements. Error bars represent the ±
one standard deviation for two independent adsorption experiments. Protein and activity
measurements for each adsorption experiment represent an average of two sample
measurements.
6.3.5 - Modeling of HRP adsorption on cellulosic fibers
Binding models that can accurately reproduce the experimental adsorption isotherms of
Figure 6.8 are required to quantify the relative binding ability of HRP on sized cellulosic
fibers. Langmuir’s and Freundlich’s empirical models developed for homogeneous and
heterogeneous gas adsorption, respectively, have been used by researchers to gain
-100
-80
-60
-40
-20
0
20
40
0.0 0.1 0.2 0.3 0.4 0.5
HR
P in
acti
vati
on [
%]
HRP equilibrium concentration [mg/ml]
Untreated
0.8% rosin-alum sizing (new)
1.6% rosin-alum sizing (new)
0.3% AKD sizing
— 127 —
insights about possible protein adsorption mechanisms [ 220 , 221 ]. Although the
application of these models is highly questionable because their assumptions are not valid
for complex sorbate-sorbent systems such as the enzyme-fiber system, the models are
still employed for proteins because they are simple and they tend to produce good
regressions as well as they can provide some idea about the surface properties relevant in
protein adsorption [222].
Therefore, the experimental adsorption isotherms of HRP on cellulosic fibers with an
increasing sizing level were fitted with both Langmuir’s and Freundlich’s empirical
models using a nonlinear regression package (SigmaPlot 2001, Version 7.101). Figures
6.10A, B, C, and D illustrate the results of these regressions. Appendix A.13 is a
compilation of the modeling outputs obtained with the software.
— 128 —
0
0.04
0.08
0.12
0.16
0 1 2 3 4 5 6 7 8 9
HR
P a
dso
rbed
[µ
mo
l/g
]
HRP equilibrium concentration [µmol/l]
Cellulosic fibers sized
with 0.8wt-% rosin
Figure 6.10. Experimental [○] HRP adsorption isotherms on increasingly sized cellulosic fibers fitted with Langmuir [__] and
Freundlich [...] empirical models. A. Untreated, B. 0.8wt-% rosin-sized C. 1.6wt-% rosin-sized, D. 0.3wt-% AKD-sized.
0
0.04
0.08
0.12
0.16
0 1 2 3 4 5 6 7 8 9
HR
P a
dso
rbed
[µ
mo
l/g
]
HRP equilibrium concentration [µmol/l]
Untreated
cellulosic fibers
0
0.04
0.08
0.12
0.16
0 1 2 3 4 5 6 7 8 9
HR
P a
ds
orb
ed
[µ
mo
l/g
]
HRP equilibrium concentration [µmol/l]
Cellulosic fibers sized
with 1.6wt-% rosin
0
0.04
0.08
0.12
0.16
0 1 2 3 4 5 6 7 8 9
HR
P a
ds
orb
ed
[µ
mo
l/g
]
HRP equilibrium concentration [µmol/l]
Cellulosic fibers sized
with 0.3wt-% AKD
A B
C D
— 129 —
Langmuir’s model was used to characterize HRP-fiber binding affinity constants (see
Table 6.4). Table 6.4 shows that the enzyme-fiber binding affinity constant increases as
the fibers become more hydrophobic, in agreement with the mechanism that enzyme-
fiber interaction is mostly governed by the hydrophobic effect. The regression quality is
lower for the untreated fibers and the AKD-treated fibers, possibly because considerable
data scattering is observed in the experimental data (a larger impact of experimental
errors on measurements of low protein concentrations). In addition, in the absence of an
adsorption plateau (untreated fibers), Langmuir’s model assumptions of a monolayer
coverage and a homogeneous distribution of the binding sites for these samples does not
seem appropriate.
Table 6.4 - Binding affinity constants, K, obtained by Langmuir’s fit of the experimental
HRP adsorption isotherms from Figure 6.6.
Fiber Sample K
[µmol/l]-1 R2
Untreated 4.1 10-5 0.50
0.8wt-% rosin 3.2 10-3 0.93
1.6wt-% rosin 1.86 10-1 0.94
0.3wt-% AKD 6.24 100 0.42
Freundlich’s model was used to characterize the adsorption heterogeneity indices of the
fibers (see Table 6.5). This index varies between 0 and 1, and the closer the value to 1,
the more homogeneous the binding sites are. Table 6.5 shows that the internal sizing
process increases the adsorption heterogeneity of the sorbents (heterogeneity indices
decrease). This observation may be explained by a nonuniform distribution of the sizing
agent on the fiber surface. The value above 1 for the heterogeneity index of the untreated
fibers cannot be explained; likely due to the inadequacy of the binding model for this
sample.
— 130 —
Table 6.5 - Heterogeneity index, m, obtained by Freundlich’s fitting of the experimental
HRP adsorption isotherms from Figure 6.6.
Fiber Sample m R2
Untreated 2.67 0.57
0.8wt-% rosin 0.87 0.93
1.6wt-% rosin 0.62 0.95
0.3wt-% AKD 0.14 0.45
The low quality of regression (R2) obtained with the AKD-sized fibers using both
Langmuir’s and Freundlich’s binding models limits the validity of the binding parameters,
making it difficult to draw a conclusion.
6.3.6 – Surface characterization of internally sized cellulosic fibers by X-ray
photoelectron spectroscopy (XPS)
Handsheets prepared with the unsized and increasingly sized cellulosic fibers used in the
previous adsorption experiments were analyzed by XPS. Figure 6.11 shows the XPS
carbon spectra of the different samples. Four distinctive carbon peaks can be observed: a
285eV binding energy peak corresponding to aliphatic primary carbon (C1) present in C-
H alkane-type carbon atoms, a 286.7eV binding energy peak attributed to secondary
carbon (C2) present in C-O alcohol/ether-type carbon atoms, a peak at 288.4eV binding
energy associated with tertiary carbon (C3) present in O-C- and C=O acetal-type carbon
atoms and a 289.2eV high binging energy peak arising from carbon atoms (C4) present in
O-C=O ester/carboxylic acid type of moiety. Table 6.6 summarizes the quantitative XPS
data corresponding to the four carbon peaks identified in the spectra for the samples.
— 131 —
a) Unsized b) 0.3wt.-% AKD-sized
c) 0.8wt.-% rosin-sized d) 1.6wt.-% rosin-sized
Figure 6.11 – X-Ray Photoelectron C1s spectra for increasingly sized handsheets. The
XPS spectra were not smoothened prior to deconvolution. A Gaussian-Lorentzian ratio of
70%/30% was used for peak deconvolution. The binding energy scale was referenced to
the C1s line of aliphatic carbon set at 285.0 eV
0
200
400
600
800
1000
1200
1400
1600
280285290295300
Cou
nts/
s
Bindind Energy [eV]
C1
C2
C3
0
500
1000
1500
2000
2500
3000
280285290295300
Cou
nts/
sBindind Energy [eV]
C1
C2C4
0
200
400
600
800
1000
1200
1400
280285290295300
Cou
nts/
s
Bindind Energy [eV]
C1C2
C3
C4
0
200
400
600
800
1000
1200
1400
280285290295300
Cou
nts/
s
Bindind Energy [eV]
C1
C2
C3
C4
— 132 —
Table 6.6 – XPS analysis of unsized and increasingly sized handsheets
Sample Surface Energy [mN/m]
O/C C1 [%]
285eV
C2 [%]
286.7eV
C3 [%]
288.4eV
C4 [%]
289.2eV
Unsized 40.51 0.59 24.31 58.66 17.03 0.00
0.8wt.-% rosin 28.40 0.48 40.93 44.91 10.97 3.19
1.6wt.-% rosin 27.65 0.42 50.36 37.33 9.08 3.23
0.3wt.-% AKD 25.25 0.17 89.46 7.04 0.00 3.49
The unsized cellulose handsheet reveals the typical C2 and C3 peaks associated with the
chemical structure of cellulose and a small C1 peak which has been previously associated
with impurities (lignin, extractives, fatty acids) [223]. Irrespective of the type of sizing
agent, all sized handsheets exhibit a significant increase in aliphatic C1 signal. The long
hydrophobic aliphatic chains present in the rosin and AKD structures explain this
increase, as expected. It can also be observed that AKD-sized fibers have almost twice as
much C1% than the rosin-sized fibers. The reason for this difference is the chemical
structure of the sizing agent molecules: AKD has two aliphatic chains, whereas rosin has
only one (See Appendix A.14). Moreover, increases in the rosin dose result in augmented
aliphatic carbon content but the relationship is not linear, suggesting that the surface of
the fibers must have reached saturation and excessively high doses of rosin (1.6wt.-%)
are not completely retained on the fibers. The conclusion is also independently supported
by the reduced gain in hydrophobicity (low decrease in the surface energy) observed in
the oversized rosin samples. In addition, a C4 signal that is not detectable in the unsized
sample becomes evident in the sized samples. This high binding energy carbon atom is
associated with the carboxyl groups of the rosin molecule and the ester bond formed
between AKD and cellulose after the covalent sizing reaction.
— 133 —
6.3.7 - Thermal stability of HRP adsorbed on cellulosic fibers
The characteristic temperature at which 50% of the enzyme is thermally
denatured is shown as a distinctive endothermic peak in the DSC thermogram
(Tm) [142]. The aim of this part of the research was to examine the changes in
HRP thermal behavior upon immobilization on cellulosic fibers with different
degrees and types of internal sizing.
Figures 6.12 and 6.13 illustrate the DSC responses of HRP in solution and after
adsorption, correspondingly. In solution, at least two small endothermic peaks
(47°C and 80°C) can be associated with the enzyme thermal unfolding (see
Figure 6.12). The observation of more than one peak might indicate the presence
of more than one isoenzyme [16-19,204] or a two-step protein denaturing
process involving melting of the tertiary structure at a low temperature followed
by removal of the heme moiety from the active site at a higher temperature
[224]. Interestingly, by comparing the thermal behavior of HRP in solution
before and after dialysis against the buffer, a third small peak (68°C) is revealed.
Likely, the additive-free dialyzed solution better displayed the thermal
denaturation of some protein fraction present in a very small amount in the
initial HRP mixture.
On one hand, after HRP is adsorbed on the cellulosic fibers, the peak located at
80°C completely disappears, regardless of the type of fiber treatment. This is
indicative of either a lower surface loading of this enzyme fraction below the
DSC detection limit, or a complete denaturation of this enzyme fraction upon
interaction with the surface. On the other hand, the HRP peak initially located at
47°C (Figure 6.12, curve D) either disappears (Figure 6.13, curve E), denoting
denaturation upon adsorption; or shifts to a lower temperature (Figure 6.13,
curve B and C), denoting a decrease in thermal stability; or shifts to higher
temperatures (Figure 6.13, curve E), denoting an increase in thermal stability.
— 134 —
Figure 6.12 - DSC control thermograms for enzyme. A) 40mM KH2PO4 buffer
pH 6.8, B) 10mg/ml HRP solution in buffer, C) Same as B after dialysis against
buffer.
From figure 6.13, it becomes apparent that cellulosic fibers hydrophobized with
rosin-based sizing agents can alter the stability of the adsorbed HRP. For
addition levels up to 0.8wt-%, only a slight decrease in Tm is observed.
However, for 1.6 wt-% dose, the active enzyme structure vanishes. These results
are in close agreement with the previous investigations reported in Chapter 5
suggesting that HRP has a lower localized enzyme activity on rosin-based
highly-sized papers. In contrast, the HRP adsorbed on AKD treated fibers shows
a slightly increase in thermal stability. The smaller area under the endothermic
peak found in this latter case is justified by a comparatively lower loading
capacity observed in Figure 6.8.
— 135 —
Figure 6.13 - DSC thermograms for wet fibers after 24h of enzyme adsorption
from 0.4mg/ml HRP solutions followed by 30 min centrifugation at 10,000 rpm
to remove supernatant. A) Blank: buffer adsorbed on untreated fiber, B) HRP
adsorbed on untreated fiber, C) HRP adsorbed on fibers treated with 0.8wt-%
rosin-based sizing, D) HRP adsorbed on fiber treated with 1.6wt-% rosin-based
sizing, E) HRP adsorbed on fiber treated with 0.3wt-% AKD sizing.
6.4 – CONCLUSIONS
In the first part, this investigation examined the adsorption behavior and
inactivation of HRP on model sorbents with various degree of hydrophobicity.
Increased adsorbed amounts and increased binding strength were correlated with
a decreased polar surface free energy component in the sorbents. However, the
main drawback exhibited by the most hydrophobic sorbent (polystyrene) was a
significant loss in enzyme activity due to surface-induced changes in the enzyme
structure. The results also suggested that for calcium carbonate hydrophilicity
— 136 —
may explain the poor bioanalytical performance of pigment coated papers
observed in Chapter 5.
In the second part, this study examined the impact of different surface charges
(positive or negative) and surface charge density (carboxyl group content) of
rayon and BKSW fibers on the HRP adsorption behavior. The impact of the
sorbent surface charge sign or the sorbent surface charge density on HRP
adsorption was relatively minor when compared to the impact of sorbent
hydrophobicity. It was found that a positive surface charge and increases in the
negative surface charge density of the sorbents, produced a decrease in the
binding affinity and loading of the enzyme per unit area. However, the oxidation
process resulted in a significant augmentation of the specific surface area of the
sorbents, improving the enzyme loading capacity per unit mass. No significant
enzyme inactivation upon adsorption on the sorbents was detected within the
experimental error.
From the first and second part of this chapter it was postulated that, as observed
in many other protein-solid systems, hydrophobic interactions are the dominant
non-covalent binding mechanism acting between HRP and the cellulosic
surfaces, while electrostatic interactions play only a minor role.
In the final part, this chapter examined the adsorption behavior and inactivation
of HRP on cellulosic fibers treated with different sizing additives. It was found
that the adsorption behavior was additive-dependent: rosin-based internal sizing
produced a considerable improvement in the enzyme adsorption, whereas AKD-
based internal sizing only slightly differentiated from the untreated fibers.
Nonetheless, as observed with the model sorbents, the HRP adsorbed amounts
and affinities consistently improved as the polar component of the surface
energy of the sorbent decreased. The method used to detect differences in
enzyme activity upon adsorption could not show evidence of loss in the
biological function of HRP. However, DSC thermograms indicated that treating
— 137 —
the fibers with doses above 0.8wt-% of a rosin-based sizing agent produced
irreversible protein unfolding upon adsorption.
In addition, XPS analysis revealed that the C1 content in the fibers increases
with sizing and is almost two times larger in the AKD-sized fibers when
compared to the rosin-sized fibers. The C1 content does not linearly increase
with rosin dose increments. This observation suggests that the fibers may have
been completely covered by the rosin at 0.8 wt.-% rosin dose and further
additions are not retained resulting in insignificant gains in hydrophobicity.
Interestingly, although AKD-sized fibers have the highest aliphatic carbon
content, the decrease in polar surface energy component attained by internally
sizing the fibers is less significant than the decrease in polar surface energy
component with rosin. It is possible that some AKD might have not reacted.
— 138 —
Chapter 7
CONCLUDING REMARKS
7.1- CONTRIBUTIONS
The major findings of this thesis were:
• Ink jet technology can be used to reliably deposit HRP on solids with a
retained activity if a suitable operational window and bio-ink formulation
are used
• The local enzyme distribution on the substrate together with the surface
energy of the substrate explain the activity of the printed HRP enzyme
• Hydrophobic interactions are the dominant interaction in HRP physical
immobilization on the cellulosic substrates. Electrostatic interactions play
a minor role
• The surface energy and the surface chemistry of the cellulosic substrates
can be engineered to enhance HRP physical immobilization by increasing
binding strength or by increasing specific surface area, correspondingly
• Cellulosic fibers are more promising supports for enzyme immobilization
in bioactive papers than pigment coatings
7.2 – SPECIFIC CONCLUSIONS
As stated in the introduction, the aim of this thesis was to address key scientific
challenges in the application of bioactive materials on paper using inkjet
technology. Based on this goal, the thesis was divided in three studies (chapters
4, 5 and 6). The specific conclusions of these studies are summarized in the next
sections.
— 139 —
7.2.1 –Ink formulation and piezoelectric inkjet printing of horseradish
peroxidase
7.2.1.1 – Effect of bio-ink additives on enzyme activity
Suitable bio-inks for inkjet printing are biologically active and reliably ejectable.
By incorporating additives to the ink formulation, the stringent surface tension
and viscosity requirements for piezoelectric inkjet jettability can be met but that
can also inactivate the enzyme. Based on the results of Chapter 4, it was found
that viscosity modifiers were the most critical additives in the HRP ink
formulation because they introduced diffusion limitations that impaired
considerably the enzyme activity. It was also discovered that modifiers with a
high efficiency in viscosity modification were most suited for bio-ink
formulations because very small amounts could substantially increase the
resulting ink viscosity with insignificant effect over the enzymatic function. The
experimental results demonstrated that CMC as a viscosity modifier exhibited
this advantage and that the non-ionic surfactant Triton X-100 reduced the
surface tension of the ink to the desired levels (30mN/m) without producing
inactivation.
7.2.1.2 - Effect of the jetting process on enzyme activity
There were some concerns about the impact of high shear rates (105s-1) on the
HRP bio-ink activity in solution when it was ejected through the 21µm-size
nozzles at high speed (7-9m/s) during inkjet printing. No evidence of adverse
effects of the jetting process on the bioactivity of the ink under the high shear
rates used in this study was found. Also, suitable printing process parameters
were identified that allowed stable, repetitive, and reliable jetting of the bio-ink
developed in this study.
— 140 —
7.2.2 – Fibrous materials as support for bioactive papers
7.2.2.1 - Effect of the paper support on the bioanalytical performance of printed
HRP
The H2O2 colorimetric sensing performance of various HRP-printed papers was
measured using the chromogenic HRP substrate ABTS. The papers included
recycled and virgin fibers, mechanical and wood free furnishes, and coated and
uncoated grades. It was demonstrated that the type of support where HRP was
immobilized affected considerably the performance of paper as a sensing device.
In qualitative analysis, different papers produced distinct color profiles and color
intensities when used under similar testing conditions. In quantitative analysis,
the performance of bioactive papers differed in sensitivity and range of
detection.
From the experimental observations of Chapter 5, it was concluded that the best
overall response pertained to a well-sized printing-grade uncoated wood-free
paper. It became apparent that paper surface energy played a role in the
bioanalytical performance of the papers. In addition, the results suggested that
papers with less purity in their fiber furnish, such as mechanical papers and
recycled papers, produced unexpected color shifts, probably due to interference
in the enzymatic reaction. Interestingly, it was observed that coated papers
produced particularly poor signal intensities; the more hydrophilic the coating,
the less intense the color response was.
In attempting to explain the discrepancies in behaviour of bioactive papers with
different cellulosic fibrous supports it was postulated that two important aspects
that can affect the bioanalytical response are the ink distribution within the paper
structure and the paper-enzyme interactions.
— 141 —
7.2.2.3 – Effect of the paper support on the spatial distribution and activity of
printed HRP enzyme
A new method using confocal scanning fluorescent microscopy was developed
that can qualitatively visualize the active enzyme distribution in paper based on
the red fluorescence selectively developed by the substrate Amplex Red upon
reaction with H2O2 in the presence of HRP as the catalyst. Using the method, the
active enzyme distribution in five paper substrates after inkjet deposition and
fluorescence development was mapped. The new technique unveiled the extent
of the spreading and penetration of the HRP bio-ink in the different types of
papers. CLSM images suggested that partial penetration of the bio-ink and
minimum spreading favours the bioanalytical response. The results
demonstrated that the activity of HRP immobilized on paper depended on the
local enzyme concentration.
The enzyme distributions observed in coated papers and thick coating layers
showed that HRP preferentially locates in the fiber cell wall and not near the
pigments or fillers. It was suggested that the dynamic microvoids in the fiber
cell wall when wet help entrap the enzyme upon drying and present a more
suitable microenvironment for the preservation of the HRP biological
functionality than pigments. As a consequence, pigment coatings seem not to
favorably contribute to the bioanalytical response. The need for a better
understanding of enzyme-fiber and fiber-pigment interactions that can explain
the differences in local enzyme activity in coated and uncoated papers became
apparent.
In addition, the CLSM method was used to map the active enzyme distribution
in increasingly sized paper handsheets. It was found that although sizing
controlled spreading and penetration of the bio-ink maximizing the local enzyme
concentration, oversizing the fibers did not reduce further the enzyme
distribution and could partially inactivate the HRP enzyme. It was concluded
that the bio-ink spatial distribution does not provide a complete explanation for
— 142 —
the different bioanalytical behaviour of the papers, because papers with a similar
ink distribution still show variations in their fluorescent responses. It was
postulated that paper-enzyme interactions can affect the localized enzyme
activity, and in turn, the sensing ability of the bioactive papers.
7.2.3 – Paper-enzyme interactions
7.2.3.1- Fundamental study of the impact of cellulosic immobilization supports on
the adsorption behaviour of HRP
Due to the amphiphilic nature of the proteins, it is generally observed that
proteins interact with surfaces mainly through hydrophobic and ionic
interactions. However, how cellulosic fiber supports interact with enzymes
during physical immobilization is largely unknown based on the literature
findings. On one hand, cellulose fibers can have different levels of
hydrophobicity due to the additives used in the papermaking process (sizing
agents). On the other hand, due to the large amount of hydroxyl groups (and
minor amount of carboxyl groups) exposed in the surface, cellulosic fibers
typically exhibit a negative surface charge. Thus, enzyme molecules can
potentially interact with the paper fibers through both hydrophobic and
electrostatic interactions. Additionally, the existence of both external and
internal surfaces within the cellulosic fibers renders the interactions between the
biomolecules and the support highly complex. Moreover, it is not clear how the
different paper-enzyme interactions affect the activity of the enzyme.
Therefore, the physical immobilization behavior of horseradish peroxidase on
various model sorbents, including both simple and non-porous substrates (such
as microcrystalline cellulose, positively and negatively charged regenerated
cellulose, oxidized rayon, ground calcium carbonate, SB latex, and polystyrene)
and complex, porous and swellable bleached kraft softwood (BKSP) pulp fibers
with varying degrees of hydrophobicity and oxidation were investigated in
Chapter 6.
— 143 —
7.2.3.2 – Impact of hydrophobicity on HRP adsorption behavior on model supports
By comparing HRP adsorption on simple non-porous model sorbents with
various degrees of hydrophobicity, it was confirmed that HRP adsorbed more
strongly and in larger quantities on more hydrophobic sorbents. Higher enzyme
binding was consistently correlated with less polar surface free energy
components in the sorbents. It was concluded that the hydrophobic effect was a
dominant type of enzyme-sorbent interaction.
By comparing enzyme inactivation isotherms on the same sorbents, it was found
that very strong hydrophobic interactions produced a considerable surface-
induced inactivation in the enzyme. Therefore, the more hydrophobic sorbents
resulted in larger amounts of adsorbed enzyme at the expense of a decrease in
local enzyme activity.
It was also verified that ground calcium carbonate, a pigment widely used in
paper coating formulations, exhibited a remarkably lower adsorption affinity
when compared to both regenerated and microcrystalline cellulose. This finding
combined with the HRP spatial distribution observed in the coating layers led to
the final conclusion that the poor sensing ability of the pigment coated papers is
mainly due to the very low local enzyme concentration.
7.2.3.3 – Impact of surface charge sign and density on HRP adsorption behavior on
model supports
By comparing HRP adsorption on both positively and negatively charged rayon and on
both rayon and cellulosic fibers sorbents with an increased negative surface charge
density obtained by TEMPO-mediated oxidation, it was confirmed that electrostatic
interactions play a minor role in HRP adsorption. By comparing enzyme inactivation
isotherms on the same sorbents, it was found that no detectable enzyme inactivation was
present.
After characterization of the adsorption behaviour of HRP on paper-related
model sorbents with different surface energies, surface charge signs, and surface
— 144 —
charge densities, it was concluded that the hydrophobic effect was the dominant
interfacial interaction between the enzyme and the cellulosic model surfaces.
Electrostatic charge interactions had a relatively minor role in the adsorption of
the enzyme molecule.
7.2.3.4 – Impact of internal sizing on HRP adsorption behavior on cellulosic fibers
The adsorption and inactivation isotherms of HRP on BKSW cellulosic pulp
fibers increasingly hydrophobized with rosin-based sizing agent and an AKD-
based sizing agent were characterized in Chapter 6. As with the model sorbents,
the more hydrophobic fibers adsorbed more enzyme. By comparing their HRP
adsorption isotherms, it was concluded that internally sizing the cellulosic fibers
is a suitable fiber treatment to achieve a stronger and greater HRP physical
immobilization. Experimental results showed that paper fibers with a similar
hydrophobicity (as measured by Hercules sizing test) sized by two different
sizing agents, had very different adsorption responses. Upon adsorption, the
rosin-sized fibers adsorbed more enzyme than the AKD-sized fibers, in direct
correlation with their lower and higher polar surface free energy components,
respectively.
Langmuir’s and Freundlich’s empirical models were fitted to the experimental
adsorption isotherms and the corresponding binding parameters were obtained.
It was concluded that rosin-sizing produced a several order of magnitude
increase in the binding affinity constants. XPS analysis has shown that
increasingly sizing the fibers results in a greater aliphatic carbon content, as
expected. However, due to the different chemical structure of the sizing agents,
AKD sized fibers have a larger C1 content than the rosin-sized fibers. No linear
correlation between the rosin dose and the C1 content was found, suggesting
that the fiber surface might be saturated with rosin and further addition does not
lead to further gains in hydrophobicity.
Inactivation isotherms did not provide conclusive evidence of HRP inactivation
upon adsorption onto highly sized cellulosic fibers. However, upon analyzing
— 145 —
the thermal behavior of HRP adsorbed on increasingly sized fibers using DSC,
HRP thermal unfolding upon adsorption onto highly rosin-sized cellulosic fibers
was demonstrated.
7.3- RECOMMENDATIONS
• It would be worthwhile to extend the systematic approach presented in this thesis
on both bio-ink formulation and inkjet deposition to other types of biomolecules
such as antibodies, phages or DNA aptamers. Moreover, the challenges involved
in printing more complex inks containing, for example, multiple bioagents or
biomolecules anchored on particles or interactive components should be
addressed.
• It would be of interest to explore and develop methods that can provide direct
characterization of the structure and distribution of biomolecules immobilized on
solids with increased spatial resolution. Furthermore, methods that could capture
the dynamic changes in the biomolecule conformation upon immobilization will
be valuable tools for interrogation of the bioactive paper structure. Spectroscopic
techniques like TOF-SIMS using new type of ion sources specially adapted for
the examination of fragile biological structures and emerging scanning probe
microscopies that rely on active probes (i.e., chemical, electrochemical, thermal)
should be assessed.
• It may be possible to implement applications of HRP-printed papers for rapid
industrial monitoring of oxidant residuals (e.g., residual peroxide content in pulp
bleaching operations, reducible sulfur activity in papers) and antioxidants (e.g.,
antioxidant activity in food and beverages).
• In general, biomolecules are available in small quantities because extraction and
purification processes are laborious and costly. It would be advantageous to
elucidate the biomolecule purity requirements for bioactive paper applications.
• Further work should be conducted to determine the fiber types best suited
for bioactive paper production. This study identified purity and
crystallinity as factors related to fiber composition that merit more
— 146 —
investigation. The impact of the differences in hemicellulose content of
both wood and non-wood fibers should also be evaluated.
— 147 —
REFERENCES
[1]. Kinnunen L, Potential of bioactive paper explored, Paperi ja Puu, 88(2):84-85+66
(2006).
[2]. Orzechowska A, Sentinel: Safety through new super papers, Pulp and Paper Canada,
107(12):26-29 (2006).
[3]. Muyrong A, Super papers, the next big thing in health and safety, Paper Asia
23(6):18-19 (2007).
[4]. McCormick C, The bioactive paper chase, Pulp and Paper Canada 109(1):17-20
(2008).
[5]. Sentinel Bioactive Paper Network, Retrieved February 12, 2009 from
http://www.bioactivepaper.ca
[6]. Aikio S, Grönqvist S, Hakola L, Hurme E, Jussila S, Kaukoniemi O, Kopola H,
Känsäkoski M, Leinonen M, Lippo S, Mahlberg R, Peltonen S, Qvintus-Leino P,
Rajamäki T, Ristchkoff AC, Smolander M, Vartianien J, Viikari L and Vilman M,
Bioactive papers and fiber products: Patent and literature survey, Oulu: Julkaisija-
Utgivare, VTT working papers 51 (2006).
[7] Lyron Z, Fischer M and Bromberg A, Novel Approaches in Biosensors and Rapid
Diagnostic Assays. New York: Kluwer Academic/Plenum Publishers, Chapter 13, p173.
(2001).
[8]. Freitag R., Utilization of Enzyme-Substrate Interactions in Analytical Chemistry,
Biomedical applications 722(1-2):279 (1999).
[9]. Byfield MP, Abuknesha RA, Biochemical aspects of biosensors, Biosensors and
Bioelectronics 9(4-5):373-399 (1994).
[10]. Sadana A, Biocatalysis : Fundamentals of Enzyme Deactivation Kinetics.
Englewood Cliffs, N.J.: Prentice Hall, Chapter 1, p1, (1991).
[11]. Niemeyer C and Mirkin C, Nanobiotechnology: Concepts, Applications and
Perspectives, Weinheim: Wiley-VCH, Chapter 3, p31, (2004).
— 148 —
[12]. Walsh G, Proteins: Biochemistry and Biotechnology, Chichester, New York: J.
Wiley, Chapter 9, p349, (2002).
[13]. Guisan J, Immobilization of Enzymes and Cells, Totowa, N.J.: Humana Press,
Chapter 1, p1, and chapter 19, p153, (2006).
[14 ]. Bernard A, Delamarche E, Schmid H, Michel B, Bosshard HR, Biebuyck H,
Printing patterns of proteins, Langmuir 14(9):2225-2229, (1998).
[15]. Schomburg, D., Salzmann, M. and Stephan, D. (Eds.). Enzyme Handbook 7. Class
1.5-1.12: Oxidoreductases. New York: Springer-Verlag, (1990).
[16]. Veitch NC. Horseradish peroxidase: A modern view of a classic enzyme.
Phytochemistry, 65(3):249-259, (2004).
[17]. Shannon,L., Kay, E. and Lew, J. Peroxides isoenzymes from horseradish roots: I –
Isolation and physical properties. The Journal of Biological Chemistry 241(9): 2166-2172,
(1966).
[18]. Kay, E., Shannon,L. and Lew, J. Peroxides isoenzymes from horseradish roots: II –
Catalytic properties, The Journal of Biological Chemistry 242(10): 2470-2473, (1967).
[19]. Kay, E., Shannon,L. and Lew, J. Peroxides isoenzymes from horseradish roots: III –
Circular dichroism of isoenzymes and apoisoenzymes, The Journal of Biological
Chemistry 243(13): 2560-2565, (1968).
[20]. Gajhede M, Schuller D, Henriksen A, Smith A, Poulos T. Crystal Structure of
Horseradish Peroxidase C at 2.15 Angstrom Resolution. Nature structural biology,
4(12):1032, (1997).
[21] Yan, M., Ge, J., Liu, Z. and Ouyang, P. Encapsulation of Single Enzyme in Nanogel
with Enhanced Biocatalytic Activity and Stability. Journal of the American chemical
Society 2006, 128: 11008-11009 (2006).
[22]. Methods of Enzymatic Analysis. Vol.1 Fundamentals, pgs. 495, 687. 3rd ed.
Weinheim: Verlag Chemie, (1983).
[23]. Dunford, H.Brian. Heme peroxidases. New York: John Wiley, (1999).
— 149 —
[24] . Ryan O, Smyth MR, Fágáin CO. Horseradish peroxidase: The analyst's friend.
Essays in biochemistry, 28:129-146, (1994).
[25]. Saunders BC. Peroxidase; the Properties and Uses of a Versatile Enzyme and of
some Related Catalysts. London: Butterworths, pgs. 34-37 (1964).
[26]. Peroxidases in Chemistry and Biology. Vol II. Chapter 1, p1. Boca Raton, Fla.:
CRC Press, (1991).
[27]. Tischer W, Kasche V. Immobilized enzymes: Crystals or carriers? Trends in
Biotechnology 17(8):326-335 (1999).
[28]. Angenendt P, Glokler J, Murphy D, Lehrach H, Cahill DJ, Toward optimized
antibody microarrays: A comparison of current microarray support materials, Analytical
Biochemistry 309(2):253-260 (2002).
[29]. Berlin P, Klemm D, Jung A, Liebegott H, Rieseler R, Tiller J, Film-Forming
Aminocellulose Derivatives as Enzyme-Compatible Support Matrices for Biosensor
Developments, Cellulose 10(4):343 (2003).
[30]. Schena M, Microarray Biochip Technology, Natick, MA: Eaton Pub., Chapter 2,
p19 and chapter 3, p39. (2000).
[31]. White CA, Kennedy JF, Popular matrices for enzyme and other immobilizations,
Enzyme and Microbial Technology, 2(2):82-90 (1980).
[32]. Murtinho D, Lagoa AR, Garcia FAP, Gil MH, Cellulose derivatives membranes as
supports for immobilisation of enzymes, Cellulose 5(4):299-308 (1998).
[33]. Gemeiner P, Štefuca V, Báleš V, Biochemical engineering of biocatalysts
immobilized on cellulosic materials, Enzyme and Microbial Technology 15(7):551-566
(1993).
[34]. Gemeiner P, Polakovič D, Mislovičová D, Štefuca D, Cellulose as a (bio)affinity
carrier: Properties, design and applications, Journal of Chromatography B: Biomedical
Sciences and Applications 715(1):245-271 (1998).
— 150 —
[35]. Lilly MD, Enzymes immobilized to cellulose, Methods in Enzymology 44:46-53,
(1976).
[36]. Gemeiner P, Enzyme Engineering : Immobilized Biosystems, English ed. New York:
E. Horwood, Chapters 2 and 4, (1992).
[37]. Greyson, J., Problems and possibilities of chemistry on dry reagent carriers, The
Journal of Automatic Chemistry, 3(2) 66, (1981).
[38]. Walter B, Fundamentals of dry reagent chemistries: The role of enzymes, Methods
of biochemical analysis 36:35-62, (1992).
[39]. Walter B, Construction of dry reagent chemistries: Use of reagent immobilization
and compartmentalization techniques, Methods in Enzymology 137:394-420, (1988).
[40]. A.W. Martinez, S.T. Phillips, M.J. Butte, G. M. Whitesides, Patterned Paper as a
Platform for Inexpensive, Low-Volume, Portable Bioassays, Angew. Chem. Int. Ed. 46,
1318, (2007).
[41]. Bruzewicz, D., Reches, M. and Whitesides G., Low-Cost Printing of
Poly(dimethylsiloxane) Barriers To Define Microchannels in Paper, Anal. Chem.
80:3387-3392, (2008).
[42]. Martinez, A., Phillips, S., Carrilho, E., Thomas, S., Sindi, H. and Whitesides, G.,
Simple Telemedicine for Developing Regions: Camera Phones and Paper-Based
Microfluidic Devices for Real-Time, Off-Site Diagnosis, Anal. Chem., 80 (10): 3699–
3707, (2008).
[43]. Li, X., Tian, J., Nguyen, T. and Shen, W., Paper-Based Microfluidic Devices by
Plasma Treatment, Analytical Chemistry, D.O.I. 10.1021/ac801729t, (2008).
[ 44 ]. Abe, K., Susuki, K. and Citterio, D., Inkjet-Printed Microfluidic Multianalyte
Chemical Sensing Paper, Analytical Chemistry, 80:6928-6934, (2008).
[45]. Barbulovic-Nad, I., Lucente, M., Sun, Y., Zhang, M., Wheeler, A. and Bussmann,
M., Bio-Microarray Fabrication Techniques—A Review, Critical Reviews in
Biotechnology, 26:237–259, (2006).
— 151 —
[46]. Nestorson, A., Jansson, A., Järnström, L., Jonssön, L., Leufvén, A., Neoh, K. and
Kang, E., Latex dispersions as carriers for glucose oxidase scavenging systems,
Proceedings of the 10th Advanced Fundamentals Coating Symposium, Montreal, Canada,
456-467, (2008).
[47]. Schena M, Protein Microarrays, Boston: Jones and Bartlett, (2005).
[48]. Zhang, X, Screen-printing methods for biosensor production, in Biosensors, Cooper
J. and Cass, A., Eds., Oxford University Press, Second Edition, pgs. 41-57, (2004).
[49]. Shalon D, Smith SJ, Brown PO, A DNA microarray system for analyzing complex
DNA samples using two-color fluorescent probe hybridization, Genome Research
6(7):639-645, (1996).
[50]. Rose D, Microdispensing technologies in drug discovery, Drug Discovery Today
4(9):411-419, (1999).
[51]. Mossoba M, Printing Microarrays of Bacteria for Identification by Infrared
Microspectroscopy, Vibrational spectroscopy 38(1):229, (2005).
[52]. Kumar A, Biebuyck H, Whitesides G, Patterning self-assembled monolayers:
Applications in materials science, Langmuir 10:1498-1511, (1994).
[53]. Xia Y, Whitesides GM, Soft lithography, Annual Review of Materials Science
28(1):153-184, (1998).
[54]. Xia Y, Whitesides GM, Soft lithography, Angewandte Chemie - International
Edition 37(5):551-575, (1998).
[55]. Kane RS, Takayama S, Ostuni E, Ingber DE, Whitesides GM, Patterning proteins
and cells using soft lithography, Biomaterials 20(23-24):2363-2376, (1999).
[56]. Michel, B., Bernard, A., Bietsch, A., Delamarche, E., Geissler, M., Juncker, D.,
Kind, H., Renault, J.-P., Rothuizen, H., Schmid, H., Schmidt-Winkel, P., Stutz, R., and
Wolf, H., Printing meets lithography: Soft approaches to high-resolution patterning,
IBM Journal of Research and Development, 45(5):697-719, (2001).
— 152 —
[57]. Csucs G, Michel R, Lussi JW, Textor M, Danuser G, Microcontact printing of novel
co-polymers in combination with proteins for cell-biological applications, Biomaterials
24(10):1713-1720, (2003).
[58]. James, C., Davis, R., Kam, L., Craighead, H., Isaacson, M., Turner, J. and Shain,
W., Patterned protein layers on solid substrates by thin stamp microcontact printing,
Langmuir 14(4):741-744, (1998).
[59]. Blawas AS, Reichert WM, Protein patterning, Biomaterials 19(7-9):595-609,
(1998).
[60]. Pritchard DJ, Morgan H, Cooper JM, Patterning and regeneration of surfaces with
antibodies, Analytical Chemistry 67(19):3605-3607, (1995).
[61]. Pirrung MC, Huang CY, A general method for the spatially defined immobilization
of biomolecules on glass surfaces using "caged" biotin, Bioconjugate Chemistry
7(3):317-321, (1996).
[62]. Hengsakul M, Cass AEG, Protein patterning with a photoactivatable derivative of
biotin, Bioconjugate Chemistry 7(2):249-254, (1996).
[63]. Mooney JF, Hunt AJ, Mcintosh JR, Liberko CA, Walba DM, Rogers CT,
Patterning of functional antibodies and other proteins by photolithography of silane
monolayers, Proceedings of the National Academy of Sciences of the United States of
America 93(22):12287-12291, (1996).
[64]. Moerman R, Frank J, Marijnissen JCM, Schalkhammer TGM, Van Dedem GWK,
Miniaturized electrospraying as a technique for the production of microarrays of
reproducible micrometer-sized protein spots, Analytical Chemistry 73(10):2183-2189,
(2001).
[65]. Avseenko NV, Morozova TY, Ataullakhanov FI, Morozov VN, Immunoassay with
multicomponent protein microarrays fabricated by electrospray deposition, Analytical
Chemistry 74(5):927-933, (2002).
— 153 —
[66]. Morozov VN, Morozova TY, Electrospray deposition as a method for mass
fabrication of mono- and multicomponent microarrays of biological and biologically
active substances, Analytical Chemistry 71(15):3110-3117, (1999).
[67]. Morozov VN, Morozova TY, Electrospray deposition as a method to fabricate
functionally active protein films, Analytical Chemistry 71(7):1415-1420, (1999).
[68]. Avseenko NV, Morozova TY, Ataullakhanov FI, Morozov VN, Immobilization of
proteins in immunochemical microarrays fabricated by electrospray deposition,
Analytical Chemistry 73(24):6047-6052, (2001).
[69]. Barron JA, Rosen R, Jones-Meehan J, Spargo BJ, Belkin S, Ringeisen BR,
Biological laser printing of genetically modified escherichia coli for biosensor
applications, Biosensors and Bioelectronics 20(2):246-252, (2004).
[70]. Barron JA, Wu P, Ladouceur HD, Ringeisen BR, Biological laser printing: A novel
technique for creating heterogeneous 3-dimensional cell patterns, Biomedical
Microdevices 6(2):139-147, (2004).
[71]. Ringeisen BR, Chrisey DB, Piqué A, Young HD, Modi R, Bucaro M, Jones-
Meehan J, Spargo B, Generation of mesoscopic patterns of viable escherichia coli by
ambient laser transfer, Biomaterials 23(1):161-166, (2002).
[72]. Natarajan, S.,Katsamba, P., Miles, A., Eckman, J., Papalia, G., Rich, R., Gale, B.,
and Myszka, D., Continuous-flow microfluidic printing of proteins for array-based
applications including surface plasmon resonance imaging, Analytical Biochemistry 373:
141–146, (2008).
[73]. Le H, Progress and trends in ink-jet printing Technology Part 1,2,3,4 and 5,
Journal of Imaging Science and Technology 42(1):49-62, (1998).
[74]. Cooley P, Wallace D, Antohe B. Applications of ink-jet printing technology to
BioMEMS and microfluidic systems. Proc. SPIE conference on Microfluidics and
BioMEMS, vol. 4560, (2001).
[75]. Basaran O, Small-scale free surface flows with breakup: drop formation and
emerging applications, AIChE Journal 48(9):1842-1848, (2002).
— 154 —
[76]. Burgold J, Weise F, Fischer M, Schlingloff G, Henkel T, Albert J, Mayer G,
Schober A, Evolution and operating experiences with different drop-on-demand systems,
Macromol Rapid Commun 26(4):265-280, (2005).
[77]. Schubert U, Ink-Jet Printing of Functional Polymers and Materials: A (Future) Key
Technology in Polymer Science, Macromolecular rapid communications 26(4):237,
(2005).
[78]. Calvert P, Inkjet printing for materials and devices, Chemistry of materials
13(10):3299-3305, (2001).
[79]. de Gans B, Schubert US, Inkjet printing of polymer micro-arrays and libraries:
Instrumentation, requirements, and perspectives, Macromol Rapid Commun 24(11):659-
666, (2003).
[80]. H. R. Kang, Water-based ink-jet ink. I. Formulation, J. Imaging Science 35(3):179-
188, (1991).
[81]. H. R.Kang, Water-based ink-jet ink. II. Characterization, J. Imaging Science 35(3):
189-194, (1991).
[82]. H. R. Kang, Water-based ink-jet ink. III. Performance studies, J. Imaging Science
35(3):195-201, (1991).
[83]. Cheng K, Yang M, Chiu W, Huang C, Chang J, Ying, T, Yang Y, Ink-jet printing,
Self-assembled polyelectrolytes, and electroless plating: Low cost fabrication of circuits
on a flexible substrate at room temperature, Macromol Rapid Commun 26(4):247-264,
(2005).
[84]. Lee, E.R. , Microdrop generation , 1st edition, CRC Press, Boca Raton , pgs. 1, 15,
55, 183, 221, 231, (2003).
[85]. 2009 Fujifilm Dimatix Inc. ,Spectra piezoelectrics. Retrieved February 12, 2009
from http://www.dimatix.com/technology/spectra-piezoelectric.asp.
— 155 —
[86]. Freire, M., Ink jet printing technology (CIJ/DOD) in Digital printing for textiles,
Ujiie,H., Boca Raton : CRC Press ; Cambridge, England : Woodhead Pub., Chapter 3,
pgs.31-52, (2006).
[87]. Menzel, C., Bibl, A. and Hoisington, P., MEMS Solutions for Precision Micro-
Fluidic Dispensing Application, IS&T NIP 20 Conference, Salt Lake City, Utah, (2004).
[88]. Dong, H., Carr, W. and Morris, J., Visualization of drop-on-demand inkjet: Drop
formation and deposition, Review of Scientific Instruments, 77, 085101, (2006).
[89]. Siringhaus, H. and Shimoda, T., Inkjet printing of functional materials, MRS
Bulletin 28 (11):802-804, (2003).
[90]. Schober, A., Günther, R., Schwienhorst, A., Döring, M. and Lindemann, B.F.,
Accurate high-speed liquid handling of very small biological samples, Biotechniques
15(2):324-329, (1993).
[91]. Blanchard AP, Kaiser RJ, Hood LE, High-density oligonucleotide arrays,
Biosensors and Bioelectronics 11(6-7):687-690, (1996).
[92]. Allain L, Stratis-Cullum D, Vo-Dinh T, Investigation of microfabrication of
biological sample arrays using piezoelectric and bubble-jet printing technologies,
Analytica chimica acta 518(1-2):77, (2004).
[93]. Goldmann T, Gonzalez J, DNA-Printing: Utilization of a Standard Inkjet Printer for
the Transfer of Nucleic Acids to Solid Supports, Journal of Biochemical and Biophysical
Methods 42(3):105, (2000).
[94]. Okamoto T, Suzuki T, Yamamoto N, Microarray Fabrication with Covalent
Attachment of DNA using Bubble Jet Technology, Nature Biotechnology 18(4):438,
(2000).
[95]. Allain L, Askari M, Stokes D, Microarray sampling-platform fabrication using
bubble-jet technology for a biochip system, Fresenius' Journal of Analytical Chemistry
371(2):146-150, (2001).
— 156 —
[96]. Hughes, T., Mao, M., Jones, A., Burchard, J., Marton, M., Shannon, K., Lefkowitz,
S., Ziman, M., Schelter, J., Meyer, M., Kobayashi, S.,1, Davis, C., Dai, H., He, Y.,
Stephaniants, S., Cavet, G., Walker, W., West, A., Coffey, E., Shoemaker, D., Stoughton,
R., Blanchard, A., Friend, S., and Linsley, P., Expression profiling using microarrays
fabricated by an ink-jet oligonucleotide synthesizer, Nature Biotechnology 19: 342-347,
(2001).
[97]. Wilson JWC, Boland T, Cell and organ printing 1: Protein and cell printers, The
Anatomical Record Part A 272A(2):491-496, (2003).
[98]. Nakamura, M., Kobayashi, A., Takagi, F., Watanabe, A., Himura, Y., Ohuchi, K.,
Iwasaki, Y., Horie, M., Morita, I. amd Takatani, S., Biocompatible inkjet printing
technique for designed seeding of individual living cells, Tissue Engineering 11(11/12):
1658-1666, (2005).
[99]. Xu, T., Jin, J., Gregory, C., Hickman, J. and Boland, T., Inkjet printing of viable
mammalian cells, Biomaterials 26:93–99, (2005).
[100]. Xu T, Petridou S, Lee E, Roth E, Vyavahare N, Hickman J, Boland T,
Construction of high-density bacterial colony arrays and patterns by the ink-jet method,
Biotechnology and Bioengineering 85(1):29, (2004).
[101]. Flickinger, M., Schottel, J., Bond, D., Aksan,A. and Scriven, L., Painting and
Printing Living Bacteria: Engineering Nanoporous Biocatalytic Coatings to Preserve
Microbial Viability and Intensify Reactivity, Biotechnol. Prog., 23:2-17, (2007).
[102]. Lonini, L., Accoto, D., Petroni S. and Guglielmelli, E., Dispensing an enzyme-
conjugated solution into an ELISA plate by adapting ink-jet printers, J. Biochem.
Biophys. Methods 70: 1180–1184, (2008).
[103]. Bae, Y., Oh, B., Lee, W., Lee, W.H. and Choi, J.W., Immunosensor for detection
of yersinia enterocolitica based on imaging ellipsometry, Anal. Chem. 76:1799-1803,
(2004).
[104]. Kido, H., Maquieira, A. and Hammock B., Disc-based immunoassays microarrays,
Analytica Chimica Acta 411:1-11, (2000).
— 157 —
[105]. Campbell, P., Miller, E., Fisher, G., Walker, L. and Weiss, L., Engineered spatial
patterns of FGF-2 immobilized on fibrin direct cell organization, Biomaterials 26:6762-
6770, (2005).
[106].Kimura J, Y. Kawana, T. Kuriyama, An Immobilized Enzyme Membrane
Fabrication Method using an Ink Jet Nozzle, Biosensors, 4, 41, (1988).
[107]. Newman JD, Turner APF, Marrazza G, Ink-jet printing for the fabrication of
amperometric glucose biosensors, Analytica chimica acta 262(1):13, (1992).
[108]. Roda A, Guardigli M, Russo C, Pasini P, Baraldini M, Protein microdeposition
using a conventional ink-jet printer, Biotechniques 28(3):492-496, (2000).
[109]. Turcu F, Hartwich G, Schäfer D, Schuhmann W, Ink-jet microdispensing for the
formation of gradients of immobilised enzyme activity, Macromol Rapid Commun
26(4):325-330, (2005).
[110]. Setti L, Fraleoni-Morgera A, Ballarin B, Filippini A, Frascaro D, Piana C, An
Amperometric Glucose Biosensor Prototype Fabricated by Thermal Inkjet Printing,
Biosensors bioelectronics 20(10):2019, (2005).
[111]. L. Setti, C. Oiana, S. Bonazzi, B. Ballarin, D. Frascaro, A. Fraleoni-Morgera, S.
Giuliani, Thermal Inkjet Technology for the Microdeposition of Biological Molecules as
a Viable Route for the Realization of Biosensors, Anal. Lett., 37, 1559, (2004).
[112]. L. Setti, A. Fraleoni-Morgera, I. Mencarelli, A. Filippini, B. Ballarin, M. Di Biase,
An HRP amperometric biosensor fabricated by thermal inkjet printing, Sensors and
Actuators B 126:252, (2007).
[113]. Mosbach, M., Zimmermann, H., Laurell, T., Nilsson, J., Csöregi, E. and
Schuhmann, W., Picodroplet-deposition of enzymes on functionalized self-assembled
monolayers as a basis for miniaturized multi-sensor structures, Biosensors &
Bioelectronics 16: 827-837, (2001).
[ 114 ]. Pardo, L., Wilson, C. and Boland T., Characterization of Patterned Self-
Assembled Monolayers and Protein Arrays Generated by the Ink-Jet Method, Langmuir,
19:1462-1466, (2003).
— 158 —
[115]. de Gans B, Duineveld P, Schubert U, Inkjet printing of polymers: State of the art
and future developments, Adv Mater 16(3):203-213, (2004).
[116]. Hart AL, Turner APF, Hopcroft D, On the use of Screen- and Ink-Jet Printing to
Produce Amperometric Enzyme Electrodes for Lactate, Biosensors Bioelectronics
11(3):263, (1996).
[117]. Lemmo AV, Fisher JT, Mario Geysen H, Rose DJ, Characterization of an inkjet
chemical microdispenser for combinatorial library synthesis, Analytical Chemistry
69(4):543-551, (1997).
[118]. Miliotis, T., Kjellström,S., Nilsson,J., Laurell, T., Edholm, L.E. and Marko-Varga,
G., Capillary liquid chromatography interfaced to matrix-assisted laser
desorption/ionization time-of-flight mass spectrometry using an on-line coupled
piezoelectric flow-through microdispenser, J. Mass Spectrom. 35: 369–377, (2000).
[119]. Sloane, A., Duff, J., Wilson, N., Gandhi, P., Hill, C., Hopwood, F., Smith, P.,
Thomas, M., Cole, R., Packer, N., Breen, E., Cooley, P., Wallace, D., Williams, K.,
Gooley, A., High Throughput Peptide Mass Fingerprinting and Protein Macroarray
Analysis Using Chemical Printing Strategies, Molecular & Cellular Proteomics 1.7: 490-
499, (2002).
[120]. Önnerfjord, P., Nilsson, J., Wallman,L., Laurell, T., and Marko-Varga, G.,
Picoliter Sample Preparation in MALDI-TOF MS Using a Micromachined Silicon Flow-
Through Dispenser, Anal. Chem., 70:4755-4760, (1998).
[121]. Antohe, B., Cooley, P., In situ synthesis of peptide microarrays using ink-jet
microdispensing, Methods in Molecular Biology 381: 299-312, (2007).
[122]. Zhang, R., Liberski, A., Khan, F., Diaz-Mochon, J., and Bradley, M., Inkjet
fabrication of hydrogel microarrays using in situ nanolitre-scale polymerisation, Chem.
Commun., 1317–1319, (2008).
[123]. Xu, T., Gregory, C., Molnar, P., Cui, X., Jalota, S., Bhaduri, S., Boland, T.,
Viability and electrophysiology of neural cell structures generated by the inkjet printing
method, Biomaterials 27:3580–3588, (2006).
— 159 —
[124]. Boland, T., Xu, T., Damon, B. and Cui, X., Application of inkjet printing to tissue
engineering, Biotechnol. J., 1: 910–917, (2006).
[ 125 ]. Tekin E, Holder E, Marin V, de Gans B, Schubert US, Ink-jet printing of
luminescent ruthenium- and iridium-containing polymers for applications in light-
emitting devices, Macromol Rapid Commun 26(4):293-297, (2005).
[126]. Hayes, D., Wallace, D. and Cox, R., MicroJet Printing of Solder and Polymers for
Multi-Chip Modules and Chip-Scale Packages, The 32nd International Symposium on
Microelectronics, International Microelectronics And Packaging Society (IMAPS)
Conference, (1999).
[127]. Scott, J., Digital printing for printed circuit boards, Circuit World 31(4): 34–41,
(2005).
[128]. Katchalski-Katzir E, Immobilized enzymes — learning from past successes and
failures, Trends in Biotechnology 11(11):471-478, (1993).
[129]. Zaborsky O, Immobilized Enzymes, Cleveland CRC Press, pgs. 1-3, (1973).
[130]. Sheldon, R., Enzyme immobilization: the quest for optimum performance, Adv.
Synth. Catal. 349: 1289-1307, (2007).
[131]. Rosevear A, Cabral J, and Kennedy J, Immobilised Enzymes and Cells, Bristol:
IOP Publishing, pgs. 4-14,(1987).
[132]. Rusmini, F., Zhong, Z. and Feijen, J., Protein immobilization strategies for protein
biochips, Biomacromolecules 8:1775-1789, (2007).
[133]. Choi M, Progress in Enzyme-Based Biosensors using Optical Transducers,
Mikrochimica acta 148(3-4):107, (2004).
[134]. Eggins BR, Chemical Sensors and Biosensors, Chichester ; Hoboken, NJ : J.
Wiley,(a) Chapter 3, p98 , (b) Chapter 2, p12, (c) Chapter 4, p107, (2002).
[135]. Stone D., Department of Chemistry, University of Toronto, CHM1102 Notes,
Section 3a, 1998-2000.
— 160 —
[136]. Chibata I, Immobilized Enzymes, Research and Development, Tokyo : Kodansha ;
New York : Wiley, c1978. Chapter 1, p1., (c1978).
[137]. Gray, J., The interaction of proteins with solid surfaces, Current Opinion in
Structural Biology, 14:110–115, (2004).
[138]. Leckband, D. and Israelachvili, J., Intermolecular forces in biology, Quarterly
Reviews of Biophysics, 34( 2):105–267, (2001).
[139]. Norde, W., My voyage of discovery to proteins in flatland ... and beyond, Colloids
and Surfaces B: Biointerfaces 61:1–9, (2008).
[140]. Hlady, V. and Buijs, J., Protein adsorption on solid surfaces, Current Opinion in
Biotechnology 7 (1):72-77, (1996).
[141]. Bhaduri, A. and Das, K., Proteins at solid water interface – A review, J.
Dispersion Science and Technolgy, 20(4),1097-1123, (1999).
[142]. Nakanishi K., Sakiyama T. and Imamura K., On the adsorption of proteins on
solid surfaces, a common but very complicated phenomenon, Journal of Bioscience and
Bioengineering, 91:233-244, (2001).
[143]. Norde W., Adsorption of proteins from a solution at the solid-liquid interface,
Advances in Colloid and Interface Science 25:267-340, (1986).
[144]. Zoungrana T., Findenegg G. and Norde W., Structure, stability and activity of
adsorbed enzymes, Journal of Colloid and Interface Science 190:437-448, (1997).
[145]. Norde W. and Zoungrana T., Surface-induced changes in the structure and activity
of enzymes physically immobilized at solid/liquid interfaces, Biotechnol. Appl. Biochem.
28:133-143, (1998).
[146]. Haynes C., Norde W., Structures and stabilities of adsorbed proteins, Journal of
Colloid and Interface Science 169:313-328, (1995).
[147]. Zhang, C, and Suslick K., Colorimetric Sensor Array for Soft Drink Analysis, J.
Agric. Food Chem., 55:237-242, (2007).
— 161 —
[148]. Medintz, I., Uyada, H., Goldman, R. and Mattoussi, H., Quantum dot
bioconjugates for imaging, labelling and sensing, Nature Materials, 4:435-446, (2005).
[149]. Zhao,W., Chiuman, W., Brook, M. and Li, Y., Simple and Rapid Colorimetric
Biosensors Based on DNA Aptamer and Noncrosslinking Gold Nanoparticle Aggregation,
ChemBioChem 8(7):727-731, (2007).
[150]. Ghauch, C. Turnar, C. Fachinger, J. Rima, A. Charef, J. Suptil, M. Martin-Bouyer,
Use of diffuse reflectance spectrometry in spot test reactions for quantitative
determination of cations in water, Chemosphere, 40:1327, (2000).
[151]. Cooper, M., Label-free screening of biomolecular interactions, Anal. Bioanal.
Chem., 377:834-842, (2003).
[152]. Knopf, J. and Bassi, A., Eds., Smart biosensor technology, CRC Press Taylor &
Francis Group, Boca Raton, pgs.8-43, (2007).
[153]. Gauglitz, G., Direct optical sensors: principles and selected applications, Anal.
Bioanal. Chem. 381:141-155, (2005).
[154]. Morgan, C., Newman, D. and Price, C., Immunosensors: technology and
opportunities in laboratory medicine, Clinical Chemistry 42(2): 193-209, (1996).
[155]. International Union of Pure and Applied Chemistry, IUPAC, Definitions:
sensitivity and linear range . Retrieved February 12, 2009 from http://goldbook.iupac.org
[ 156 ]. Ekins, R. and Edwards, P., Point on the meaning of “sensitivity”, Clinical
Chemistry 43(10):1824-1837, (1997).
[157]. Nice, E., and Catimel, B., Instrumental biosensors, new perspectives for the
analysis of biomolecular interactions, BioEssays21(4):339-352, (1999).
[ 158 ]. Turner, P., Karube, A. and Wilson, G., Eds., Biosensors Fundamentals and
Applications, New York, Oxford University Press, pgs. 85-89, (1987).
[ 159 ]. Ekins, R., Ligand assays: from electrophoresis to miniaturized microarrays,
Clinical Chemistry 44(9):2015-2030, (1998).
— 162 —
[160]. Angenendt, P., Glökler, J., Murphy, D., Lehrach, H. and Cahill, D., Toward
optimized antibody microarrays: a comparison of current microarray support materials,
Analytical Biochemistry 309:253–260, (2002).
[161]. R.H.Dettre,R.E.Johnson, Surface Properties of Polymers- I. The Surface Tensions
of Some Molten Polyethylenes, Colloid Interface Sci., 21: 367- 377, (1966).
[162]. Kitaoka, T., Isogai, A. and Onabe, F., Chemical modification of pulp fibers by
TEMPO-mediated oxidation, Nordic Pulp and Paper Research Journal, 14(4): 279-284,
(1999).
[163]. van Oss, C. J., Good, R. J. and Chaudhury, M. K., The Role of van der Waals
Forces and Hydrogen Bonds in "Hydrophobic Interactions" between Biopolymers and
Low Energy Surfaces, Journal of Colloid and Interface Science, 111(2):378-390, (1986).
[164]. Kunio Esumi Ed., Polymer Interfaces and Emulsions, New York, Marcel Dekker p.
559, (1999).
[165]. Kaelble, D. H., Physical chemistry of adhesion, New York, Wiley-Interscience,
(1971).
[166].Wu, S., Polymer Interface and Adhesion, New York, Marcel Dekker, p.180, (1982).
[ 167 ]. Felix, J. and Gatenholm, P., Controlled Interactions in Cellulose-Polymer
Composites. I: Effect on Mechanical Properties, Polymer Composites, 14(6): 449-457,
(1993).
[ 168 ]. Pasquini, D., Belgacem, M., Gandini, A., da Silva Curvelo, A., Surface
esterification of cellulose fibers: Characterization by DRIFT and contact angle
measurements, Journal of Colloid and Interface Science 295:79–83, (2006).
[ 169 ]. Jacob, P. and Berg, J., Acid-Base Surface Energy Characterization of
Microcrystalline Cellulose and Two Wood Pulp Fiber Types Using Inverse Gas
Chromatography, Langmuir 10: 3086-3093, (1994).
[ 170 ]. Belgacem, M., Blayo, A. and Gandini, A., Surface Characterization of
Polysaccharides, Lignins, Printing Ink Pigments, and Ink Fillers by Inverse Gas
Chromatography, Journal of Colloid and Interface Science 182, 431–436, (1996).
— 163 —
[ 171 ]. Dourado, F., Gama, F., Vhibowski, E. and Mota, M., Characterization of
cellulose surface free energy, J. Adhesion Sci. Technol., 12 (10): 1081–1090, (1998).
[172]. Katz, S. and Gray, D., The Adsorption of Hydrocarbons on Cellophane, Journal of
Colloid and Interface Science, 82(2): 318-325, (1981).
[ 173 ]. Aulin, C., Shchukarev, A., Lindqvist, J., Malmström, E., Wågberg, L. and
Lindström, T., Wetting kinetics of oil mixtures on fluorinated model cellulose surfaces,
Journal of Colloid and Interface Science 317:556–567, (2008).
[174]. Buschle-Diller, G., Inglesby, M. and Wu, Y., Physicochemical properties of
chemically and enzymatically modified cellulosic surfaces, Colloids and Surfaces A:
Physicochem. Eng. Aspects 260: 63–70, (2005).
[175]. Hedenberg, P. and Gatenholm, P., Conversion of Plastic/Cellulose Waste into
Composites. II. Improving Adhesion Between Polyethylene and Cellulose Using Ozone,
Journal of Applied Polymer Science, 60:2377-2385, (1996).
[176]. Al-Turaif, H., Relation between surface chemistry and surface energy of different
shape pigment blend coatings, J. Coat. Tech. Res. 5(1):85-91, (2008).
[177]. Wu, W., Giese Jr., R.F., Van Oss, C.J., Change in surface properties of solids
caused by grinding, Powder Technology, 89 (2): 129-132, (1996).
[178]. Kan, C. and Van Gilder, R., Measurement of latex surface energy and its role in
paper coating applications, 2004 TAPPI Coating and Graphic Arts Conference, Baltimore,
MD, USA, (2004).
[179]. Fowkes, F.M., Attractive Forces at Interfaces, Ind. Eng. Chem., 56(12):40-52,
(1964).
[180]. Owens, D.K., Wendt, R.C., Estimation of the Surface Free Energy of Polymers, J.
Appl. Polym. Sci. 13 :1741-1747, (1969).
[181]. van Oss, C.J., Chaudhury, M.K., Good, R.J., Interfacial Lifshitz-van der Waals
and polar interactions in macroscopic systems, Chem. Revs., 88 (6): 927-941, (1988).
— 164 —
[182]. Spelt, J.K., Li, D., The equation of state approach to interfacial tensions, in:
Neumann, A.W., Spelt, J.K., Eds. , Applied Surface Thermodynamics, Marcel Dekker
Inc.: New York, pgs.239-292, (1996).
[183]. Zisman, W.A., Contact angle, wettability and adhesion, in: Advances in Chemistry
Series, vol. 43, American Chemical Society:Washington, DC, (1964).
[184]. Kwok, D.Y., Neumann, A.W., Contact angle measurement and contact angle
interpretation, Advances in Colloid and Interface Science 81: 167-249, (1999).
[185]. Borch, J., Thermodynamics of polymer adhesion: a review, J. of Adhesion Science
and Technology, 5(7): 523-541, (1991).
[186]. van Oss, C., Chaudhury, M. and Good, R., Monopolar surfaces, Advances in
Colloid and Interface Science, 28: 35-64, (1987).
[187]. van Oss, C., Interfacial forces in aqueous media, 2nd Ed., Boca Raton, CRC Press
p. 221, 222, (2006).
[188]. Kaewparasit, C., Hequet, E., Abidi, N., and Gourlot, J.P., Application of methylene
blue adsorption to cotton fiber specific surface area measurement: Part I. Methodology,
The Journal of Cotton Science 2:164-173, (1998).
[189]. Sigma Aldrich SPABTS02.001. Enzymatic assay of peroxidase (EC 1.11.1.7) 2, 2’
– Azino-bis (3-Ethylbenzthiazoline-6-sulfonic acid) as substrate, (1996).
[190]. Askelof, P., Korsfeldt, M. and Mannervik, B., Error Structure of Enzyme Kinetic
Experiments Implications for Weighting in Regression Analysis of Experimental Data,
Eur. J. Biochem. 69: 61 -67, (1976).
[191]. Das, D., Dasgupta, A. and Kumar Das, P., Improved activity of horseradish
peroxidase (HRP) in ‘specifically designed’ ionic liquid, Tetrahedron Letters 48: 5635–
5639, (2007).
[192]. Macková, M., Ferri, E., Demnerová, K. and Macek, T., Quantitative
chemiluminescent detection of plant peroxidases using a commercial kit originally
designed for blotting assays, Chem. Listy 95: 130-132, (2001).
— 165 —
[193]. Bradford, M., A rapid and sensitive method for the quantitation of micrograms
quantities of protein utilizing the principle of protein-dye binding, Analytical
Biochemistry 72: 248-254, (1976).
[ 194 ]. Compton, S. and Jones, C., Mechanism of dye response and interference in
Bradford protein assay, Analytical Biochemistry 151: 369-374, (1985).
[195]. Childs, R. and Bardsley, W., The steady state kinetics of peroxidase with 2,2' azino
di (3 ethylbenzthiazoline 6 sulphonic acid) as chromogen, Biochemical Journal,
145(1):93, (1975).
[196]. Körtum, G., Reflectance Spectroscopy, New York, Springer, pgs.178-186, (1969).
[ 197 ]. Zhou, M., Diwu, Z., Panchuk-Voloshina, N. and Haugland, R.., A stable
nonfluorescent derivative of resorufin for the fluorometric determination of trace
hydrogen peroxide: applications in detecting the activity of phagocyte NADPH oxidase
and other oxidases, Analytical Biochemistry 253: 162, (1997).
[198]. Hudd, A. and Fox, J., Recent Advances in Inkjet Ink Technologies, IS&T's NIP 13:
International Conference on Digital Printing Technologies, 615-620, (1997).
[199]. Fu, Z., Pigmented ink formulation in Digital printing for textiles, Ujiie,H., Boca
Raton : CRC Press ; Cambridge, England : Woodhead Pub., Chapter 13, pgs. 218-232,
(2006).
[200]. Ashley C., Edds, K. and Elbert D., Development and Characterization of Ink for
an Electrostatic Ink Jet Printer, IBM Journal of Research and Development 21(1): 69-74,
(1977).
[201]. H. Dong, W.W. Carr, J.F. Morris, An experimental study of drop-on-demand drop
formation, Phys. Fluids 18, art. no. 072102, (2006).
[202]. 2009 Fujifilm Dimatix Inc, Jettable Fluid Formulation Guidelines, Retrieved
February 12, 2009 from http://www.dimatix.com/files/Dimatix-Materials-Printer-
Jettable-Fluid-Formulation-Guidelines.pdf
— 166 —
[203]. Gullichsen, J, and Paulapuro, H., Eds., Papermaking Science and Technology,
Book 13, Printing, section 6.3, (2000).
[204]. Di Risio, S. and Yan, N., Piezoelectric Ink-Jet Printing of Horseradish
Peroxidase: Effect of Ink Viscosity Modifiers on Activity, Macromol. Rapid Commun.,
28:1934, (2007).
[205]. Savelli, G., Spretib, N. and Di Profio, P., Enzyme activity and stability control by
amphiphilic self-organizing systems in aqueous solutions, Current Opinion in Colloid &
Interface Science 5:111-117, (2000).
[206]. Bruce, C., Dependence of ink jet dynamics on fluid characteristics, IBM Journal of
Research and Development, 20(3): 258-270, (1976).
[ 207 ]. B. Derham, J. Harding, The effect of the presence of globular proteins and
elongated polymers on enzyme activity, Biochim. Biophys. Acta Prot. Proteom.
1764(100):1000-1006, (2006).
[208]. R. Mukerjea, G. Slocum, J. Robyt, Significant differences in the activities of α-
amylases in the absence and presence of polyethylene glycol assayed on eight starches
solubilized by two methods, Carbohydr. Res., 341(12): 2049-2054, (2006).
[209]. A. Uribe, J. Sampedro, Measuring solution viscosity and its effect on enzyme
activity, Biol. Proced. Online, 5 (1):108-115, (2003).
[210]. U. Kästner, H. Hoffmann, R. Dönges, J. Hilbig, Structure and solution properties
of sodium carboxymethyl cellulose, Colloids and Surfaces A: Physicochemical and
Engineering Aspects 123-124: 307-328, (1997).
[211]. Sigma Aldrich, Stability of horseradish peroxidase solutions. Retrieved February
12, 2009 from http://www.sigmaaldrich.com/life-science/metabolomics/enzyme-
explorer/analytical-enzymes/peroxidase-enzymes.html
— 167 —
[212]. Chaplin M., Enzyme technology, 1st ed., Cambridge University Press, New York,
p. 80, (1992).
[213]. Alince, B., Porosity of swollen pulp fibers revisited, Nordic Pulp and Paper
Research Journal, 17(1): 71, (2002).
[214]. Sandwick R. and Schray K., Conformational states of enzymes bound to surfaces,
Journal of Colloid and Interface Science 121:1-12, (1986).
[215]. Sandwick R. and Schray K., The inactivation of enzymes upon interaction with a
hydrophobic latex surface, Journal of Colloid and Interface Science 115:130-138, (1986).
[216]. Hiner, A., Hernandez-Ruiz, J., Arnao, M., García-Cánovas, F. and Acosta, M., A
Comparative Study of the Purity, Enzyme Activity, and Inactivation by Hydrogen
Peroxide of Commercially Available Horseradish Peroxidase lsoenzymes A and C,
Biotechnology and Bioengineering 50: 655-662, (1996).
[217]. Gardner, D., Shi, S. and Tze, W., Comparison of acid-base characterization
techniques on lignocellulosic surfaces, in , Acid-Base interactions: Relevance to
adhesion science and technology, Mittal, K. Ed., VSP BV, The Netherlands, Vol. 2., p.
363, (2000).
[218]. Huang, Y., Gardner, D., Chen, M. and Biermann, C., Surface energetics and acid-
base character of sized and unsized paper handsheets, Journal of Adhesion Science and
Technology (11):1403-1411, (1995).
[219]. Hubbe, M., Paper resistance to wetting – a review of internal sizing chemicals and
their effects, BioResources 2(1): 106-145, (2006).
[220 ]. Schmitt, A.; Varoqui, R.; Uniyal, S.; Brash, J.; Pusineri, C., Interaction of
fibrinogen with solid surfaces of varying charge and hydrophobic-hydrophilic balance. I.
Adsorption isotherms, J. Colloid Interface Sci., 92(1): 25-34, (1983).
[221]. Moreno, E.; Kresak, M.; Kane, J.; Hay, D., Adsorption of proteins, peptides, and
organic acids from binary mixtures onto hydroxylapatite, Langmuir, 3: 511-519, (1987).
— 168 —
[222]. Wojciechowski, P.; Brash, A computer simulation for the study of
macromolecular adsorption with special applications to single-component protein
adsorption , J. J. Colloid Interface Sci., 140(1): 239-252, (1990).
[223]. Belgacem, M., Czeremuskin, G. , Sapieha, S. and Gandini, A., Surface
characterization of cellulose fibres by XPS and inverse gas chromatography, Cellulose
2:145-157, (1995).
[224]. Chattopadhyay K. and Mazumdar S., Structural and Conformational Stability of
Horseradish Peroxidase: Effect of Temperature and pH, Biochemistry 39263-270, (2000).
