raman spectroscopy techniques for detection of …thesis).pdf · sensitivity over current explosive...
TRANSCRIPT
RAMAN SPECTROSCOPY
TECHNIQUES FOR DETECTION OF
ENERGETIC MATERIALS AT HIGH
AND LOW CONCENTRATION
Arniza Khairani Mohd Jamil
BSc (Forensic Science) (Hons.)
A thesis submitted in fulfilment of the requirements for the degree of
Doctor of Philosophy
School of Chemistry, Physics and Mechanical Engineering
Faculty of Science and Engineering
Queensland University of Technology
2015
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration i
Statement of Original Authorship
The work contained in this thesis has not been previously submitted to meet
requirements for an award at this or any other higher education institution. To the
best of my knowledge and belief, the thesis contains no material previously
published or written by another person except where due reference is made.
QUT Verified Signature
ii Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration iii
Keywords
Surface-enhanced Raman spectroscopy (SERS); ultra-trace analysis; gold
nanoparticles; homogeneous nanostructure assembly; 2,4,6-trinitrotoluene (TNT);
Meisenheimer complex; cysteamine; 6-aminohexanethiol (AHT);
iv Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration v
Abstract
This study extends the capabilities and efficiency of surface enhanced Raman
spectroscopy (SERS) for on-site detection and identification of trace amounts of
energetic materials. The prevalence of explosive compounds in areas such as
worldwide terrorist activities, military activities as well as in industry has triggered
ongoing research for reliable, effective and efficient removal of threats from these
compounds. The long-term hazardous effects of these compounds require real time
sensing platforms capable of rapid and excellent discrimination down to very low
concentrations. The constant struggle for effective monitoring of energetic materials
provides the rationale for the present research to address the drawbacks of current
detection systems. The versatility of SERS was explored through modification of
SERS substrates in order to obtain high SERS activity and thus improve the limit of
detection of explosive threats for practical applications.
This research presents trace level detection of common explosives in aqueous and
environmental samples. Herein we demonstrate the feasibility of using colloidal gold
nanoparticles and nanostructured SERS substrates fabricated by electrodeposition
method for screening trace amounts of the explosive 2,4,6-trinitrotoluene (TNT). The
main focus is the development of efficient SERS nanosensors by fabrication of
functionalized metal surfaces. It is known that TNT molecules have the ability to
form Meisenheimer complexes. Therefore, the selective and sensitive detection of
TNT was evaluated by modifying the surfaces with a molecular attachment that is
capable of forming Meisenheimer complexes with nitroaromatic explosives.
Aminothiols, namely cysteamine and 6-aminohexanethiol (AHT) were selected as
recognition molecules because of their ability to assemble on gold by the thiol end
and interact with TNT molecules through its amine end. Our results showed better
SERS enhancements with functionalized nanostructured substrates as compared to
functionalized nanoparticles. The nanostructured substrates utilizing both cysteamine
and AHT as the recognition molecule for Meisenheimer complex formation were
able to detect TNT at a low detection limit of 100 fM. Optimizing the complex
formation at the right pH prevented cyclic binding of the aminothiols through their
amine group and thus facilitated assembly to the metal surface in an upward position.
vi Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
This contributed to improved specific binding of TNT to the amine group of both
cysteamine and 6-aminohexanethiol and consequently a stronger signal
enhancement. Selectivity for TNT was demonstrated experimentally through
colorimetric, UV-Vis and Raman studies. Both aminothiols were incapable of
forming Meisenheimer complexes with 2,4-dinitrotoluene (2,4-DNT) and picric acid
(PA). The selectivity study was critical to establish whether there would be potential
interference from molecules that commonly co-exist with TNT. In addition, as a
means to simulate real world conditions, soil samples spiked with TNT were tested.
Reliable acquisition of Raman spectrum at 1 nM of extracted TNT was obtained with
both cysteamine and AHT functionalized nanostructured substrates. The results
indicate clearly the applicability of this methodology for analysis of TNT in
environmental samples.
The proposed methodology presented in this thesis has provided an alternative
simplified approach for analysis of TNT in complex matrices with minimum
requirements for sample pre-treatment. The key parameter of this methodology is the
fabrication of a homogenous nanostructured substrate, which produced reproducible
results even at low sample concentrations. New avenues were explored to acquire
cost-effective nanostructured surfaces with increased efficiency. Due to their
preparation, the nanostructured substrates have a more homogeneous surface, which
leads to better reproducibility across the surface. The uniform distribution of hot
spots on the metal surface provided a smaller variation of SERS intensities and hence
a high SERS substrate enhancement factor (SSEF). Evidence on reproducible SERS
measurements was obtained under wide area illumination (WAI) settings, resulting in
a low relative standard deviation of 4.37%. In addition, these settings are compatible
with handheld Raman devices, encouraging application on-site.
The careful selection of suitable recognition molecule aided the sensitive and
selective targeting of the desired molecule. The optimization of the SERS substrate
and the attached chemical moiety contributed to significant improvements on
sensitivity over current explosive detection methods. The proposed technique has
given a new dimension for forensic investigation and environmental applications.
Clear discrimination and low detection limit with added reproducibility demonstrate
the strong potential of this nanosensor for field detection of various trace explosives.
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration vii
List of Publications
Published journal articles
A.K.M. Jamil, E.L. Izake, A. Sivanesan, P.M. Fredericks, Rapid detection of TNT in
aqueous media by selective label free surface enhanced Raman spectroscopy,
Talanta, 2015. 134: p 732-738
A.K.M. Jamil, E.L. Izake, A. Sivanesan, R. Agoston, G.A. Ayoko, A homogeneous
surface-enhanced Raman scattering platform for ultra-trace detection of
trinitrotoluene in the environment, Analytical Methods, 2015. 7(9): p 3863-3868
A.K.M. Jamil, A. Sivanesan, E.L. Izake, G.A. Ayoko, P.M. Fredericks, Molecular
recognition of 2,4,6-trinitrotoluene by 6-aminohexanethiol and surface-enhanced
Raman scattering sensor, Sensors and Actuators B: Chemical, 2015. 221: p 273-280
Presentation
Surface-enhanced Raman spectroscopy for trace detection of TNT, 3rd
Nanotechnology and Molecular Science HDR Symposium, 12-13 February 2015,
Queensland, Australia, ORAL
viii Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration ix
Table of Contents
Statement of Original Authorship ................................................................................. i
Keywords .................................................................................................................... iii
Abstract ........................................................................................................................ v
List of Publications .................................................................................................... vii
Table of Contents ........................................................................................................ ix
List of Figures ........................................................................................................... xiii
List of Tables............................................................................................................. xvi
List of Abbreviations................................................................................................ xvii
Acknowledgements ................................................................................................... xix
Statement of Social Impact .......................................................................................... 1
INTRODUCTORY REMARKS ............................................................................... 3
CHAPTER 1: INTRODUCTION ............................................................................ 7
1.1 Background .......................................................................................................... 7
1.2 Current techniques for explosive detection ........................................................ 10
1.2.1 Canine detection ...................................................................................... 10
1.2.2 Ion Mobility Spectroscopy (IMS) ........................................................... 11
1.2.3 X-ray scattering ....................................................................................... 11
1.2.4 Colorimetric assays ................................................................................. 12
1.2.5 Fluorescence assays................................................................................. 12
1.2.6 Tetrahertz spectroscopy (THz) ................................................................ 13
1.3 Raman spectroscopy .......................................................................................... 13
1.4 Surface enhanced Raman spectroscopy (SERS) ................................................ 16
1.4.1 Electromagnetic enhancement................................................................. 17
1.4.2 Chemical enhancement ........................................................................... 18
1.5 SERS substrates ................................................................................................. 18
1.5.1 Gold substrates ........................................................................................ 20
x Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
1.6 Self-assembled monolayers (SAMs) .................................................................. 21
1.6.1 Preparation of SAMs ............................................................................... 24
1.7 Alkanethiols SAMs ............................................................................................ 25
1.7.1 Mixed SAMs ........................................................................................... 26
1.8 Meisenheimer complex ...................................................................................... 27
1.9 SERS detection of TNT ..................................................................................... 28
1.10 Research objectives .......................................................................................... 32
1.11 Samples and their Raman characteristics ......................................................... 33
CHAPTER 2: CYSTEAMINE-FUNCTIONALIZED NANOPARTICLES FOR
SELECTIVE SERS DETECTION OF TNT IN FORENSIC APPLICATIONS39
2.1 Preface ................................................................................................................ 43
2.2 Abstract .............................................................................................................. 44
2.3 Graphical abstract............................................................................................... 45
2.4 Introduction ........................................................................................................ 46
2.5 Experimental ...................................................................................................... 49
2.5.1 Reagents .................................................................................................. 49
2.5.2 Gold nanoparticle synthesis ..................................................................... 49
2.5.3 Preparation of Meisenheimer complex with TNT ................................... 50
2.5.4 Detection of TNT by surface enhanced Raman spectroscopy
(SERS) ..................................................................................................... 50
2.5.5 Interference study .................................................................................... 50
2.5.6 Instrumentation ........................................................................................ 51
2.6 Results and discussion ....................................................................................... 51
2.6.1 Development of Meisenheimer complex ................................................ 51
2.6.2 SERS detection of TNT ........................................................................... 55
2.6.3 Interference study .................................................................................... 60
2.7 Conclusion ......................................................................................................... 61
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration xi
CHAPTER 3: REPRODUCIBLE NANOSENSOR FOR SERS DETECTION
OF TNT IN ENVIRONMENTAL APPLICATIONS ........................................... 63
3.1 Preface ................................................................................................................ 67
3.2 Abstract .............................................................................................................. 68
3.3 Graphical abstract .............................................................................................. 69
3.4 Introduction ........................................................................................................ 70
3.5 Experimental section .......................................................................................... 72
3.5.1 Chemicals and materials.......................................................................... 72
3.5.2 Instrumentation........................................................................................ 72
3.5.3 SERS substrate preparation ..................................................................... 73
3.5.4 Preparation of cysteamine- TNT complex .............................................. 73
3.5.5 Detection of TNT by SERS ..................................................................... 74
3.5.6 Identification of TNT in soil ................................................................... 74
3.6 Results and discussion ....................................................................................... 74
3.6.1 pAu/AuNS substrate preparation and characterization ........................... 74
3.6.2 SERS detection of TNT........................................................................... 79
3.6.3 Identification of TNT in soil ................................................................... 82
3.7 Conclusion ......................................................................................................... 83
3.8 Acknowledgement ............................................................................................. 83
3.9 Supporting information ...................................................................................... 84
CHAPTER 4: AMINOHEXANETHIOL MIXED MONOLAYER
NANOSENSOR FOR SERS DETECTION OF TNT IN THE ENVIRONMENT89
4.1 Preface ................................................................................................................ 93
4.2 Abstract .............................................................................................................. 94
4.3 Graphical abstract .............................................................................................. 95
4.4 Introduction ........................................................................................................ 96
4.5 Experimental section .......................................................................................... 98
4.5.1 Chemicals and materials.......................................................................... 98
4.5.2 Instrumentation........................................................................................ 99
xii Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
4.5.3 SERS substrate preparation and mixed monolayer formation ................ 99
4.5.4 Preparation of Meisenheimer complex with TNT both in solution
and on surface ........................................................................................ 100
4.5.5 Extraction of TNT from soil .................................................................. 100
4.6 Results and Discussion ..................................................................................... 101
4.6.1 Characterization of AHT-TNT Meisenheimer complex in solution
phase ...................................................................................................... 101
4.6.2 Electrochemical and XPS characterization of CA and AHT
modified electrodes/substrates .............................................................. 105
4.6.3 Characterization of AHT-TNT Meisenheimer complex on
pAu/AuNS surface ................................................................................ 108
4.6.4 AHT and BT mixed monolayer system for efficient TNT complex
formation ............................................................................................... 110
4.6.5 Interference study .................................................................................. 113
4.6.6 Detection of TNT in soil ....................................................................... 115
4.7 Conclusion ....................................................................................................... 117
4.8 Acknowledgement............................................................................................ 117
CHAPTER 5: CONCLUSIONS ........................................................................... 119
5.1 Future work ...................................................................................................... 122
REFERENCES ....................................................................................................... 125
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration xiii
List of Figures
Figure 1.1: Energy level diagram of the various light scattering processes. ............. 15
Figure 1.2: Schematic representation of the surface plasmon electron
oscillation. Adapted from reference [45]. .................................................... 17
Figure 1.3: Schematic representation of SAMs formed from alkanethiols on a
gold substrate surface. Yellow circles represent chemisorbed thiol
head-groups whereas the green circles represent the terminal groups of
any chemical functionality. Adapted from reference [60]. .......................... 25
Figure 1.4: Schematic illustration of the possible immobilization of TNT on
mixed SAMs on a gold surface. ................................................................... 26
Figure 1.5: Nucleophilic reaction between electron-deficient nitroaromatic
compound and an amine (electron-rich nucleophile) to form a
Meisenheimer complex. ............................................................................... 27
Figure 1.6: Raman spectrum of neat TNT under 785nm excitation wavelength. ..... 34
Figure 1.7: Raman spectrum of neat DNT under 785nm excitation wavelength. .... 35
Figure 1.8: Raman spectrum of neat picric acid under 785nm excitation
wavelength. .................................................................................................. 36
Figure 2.1: Development of 1:1 cysteamine-TNT Meisenheimer complex in
aqueous medium at pH 8.5........................................................................... 52
Figure 2.2: (a) UV-Vis spectrum of the aqueous 1:1 Meisenheimer complex
between cysteamine and TNT and (b) the development of the complex
with time. ..................................................................................................... 53
Figure 2.3: Mechanism for the formation of 1:1 cysteamine-TNT
Meisenheimer complex. ............................................................................... 54
Figure 2.4: (a) TEM and (b) size distribution of unmodified gold nanoparticles ..... 56
Figure 2.5: UV-Vis absorption spectra of AuNP and AuNP-Meisenheimer
complex. ....................................................................................................... 57
xiv Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
Figure 2.6: Raman spectrum of 10-6 M TNT. The TNT solution was first
reacted with cysteamine then with unmodified AuNPs for SERS
detection. ...................................................................................................... 57
Figure 2.7: Quantification of ultra trace amounts of TNT in aqueous solution ........ 58
Figure 2.8: (a) UV-Vis absorption spectra and (b) SERS spectra of
cysteamine/TNT, cysteamine/DNT and cysteamine/picric acid
mixtures. The final TNT concentration for the UV-Vis measurements
was 10-4 M and for the SERS measurements 10-6 M. The inset depicts
the reference UV-Vis spectra of 2,4-DNT and picric acid. ......................... 61
Figure 3.1: (A-D) SEM images of pAu/AuNS surface, prepared at -400mV,
under different magnifications. .................................................................... 76
Figure 3.2: A series of SERRS spectra of 4α-CoIITAPc randomly collected
over the entire 8 mm diameter pAu/AuNS disc using (A) 50x
objective with a laser focusing area of 3.14 µ2 and (B) 5x objective
with a laser focusing area of 132.67 µ2. ....................................................... 78
Figure 3.3: (A) SERS spectra of decreasing TNT concentration: 1 nM (pink),
100 pM (blue), 10 pM (green), 1 pM (red), 100 fM (black). Inset
shows the closer view of 100 fM spectrum. For each concentration, 25
spectra were randomly recorded over the entire surface and averaged.
(B) Plot demonstrating the linear relation between the concentration of
TNT and Raman intensity @1348 cm-1. ...................................................... 81
Figure 3.4: Monolayer SERS spectra of bare pAu/AuNS (blue), 10 pM
cysteamine (green), 1 nM cysteamine-TNT (black) and 1 nM soil
extracted TNT complexed with cysteamine (red) on pAu/AuNS
surfaces. ........................................................................................................ 82
Figure 4.1: UV–Vis absorption spectra of aqueous 1:1 Meisenheimer complex
between AHT and TNT (A) at neutral and alkaline pH and (B)
Absorbance vs time plot for the formation of Meisenheimer complex
between AHT and TNT at pH 8.5. ............................................................. 102
Figure 4.2: (A) Solution Raman spectrum of 1 mM AHT at pH 8.5 (black),
TNT at pH 8.5 (blue), AHT:TNT mixture at pH 7 (green) and pH 8.5
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration xv
(red). (B) The same spectra as shown in A focused at the 2800-3000
cm-1 region. ................................................................................................ 104
Figure 4.3: Linear sweep voltammograms obtained for the self-assembled
monolayer of (a) cysteamine (b) AHT in 0.1 M KOH at a scan rate of
0.1 Vs-1. ...................................................................................................... 106
Figure 4.4: XPS spectra corresponding to N 1s region of self-assembled
monolayer of (a) cysteamine (b) AHT. ...................................................... 107
Figure 4.5: (A) SERS spectra of AHT monolayer (black) and AHT+TNT
surface bound complex on pAu/AuNS surface at pH 8.5. 1 nM TNT
concentration was used for complex formation. Inset shows the
zoomed in amine stretching region. (B) SERS intensity vs time plot
for the formation of Meisenheimer complex between AHT and TNT at
pH 8.5. ........................................................................................................ 109
Figure 4.6: SERS spectra of TNT complex on AHT monolayer (black), AHT
and BT mixed monolayer in the ratio of 1:1 (blue) and 9:1 (red). ............. 111
Figure 4.7: (A) SERS spectra @1608 cm-1 obtained for various concentrations
of TNT on AHT:BHT (9:1) modified pAu/AuNS surface at pH 8.5.
(B) Plot demonstrating the linear correlation between logarithmic
solution concentration of TNT and logarithmic SERS intensity of the
AHT-TNT complex. .................................................................................. 112
Figure 4.8: UV-Vis absorption spectra of 2,4-DNT (green), PA (brown), TNT
(blue), AHT+DNT (black), AHT+PA (red) and AHT+TNT (pink) at
pH 8.5. ........................................................................................................ 114
Figure 4.9: SERS spectra of 1 nM (black) and 2 nM soil extracted (red) TNT
on AHT:BHT (9:1) modified pAu/AuNS surface at pH 8.5. ..................... 116
Figure 5.1: A handheld Raman device that can easily be interfaced with other
platforms. Adapted from reference [226]. ................................................. 124
xvi Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
List of Tables
Table 1.1: Selected examples of common explosives. ................................................ 9
Table 1.2: Characteristic Raman modes of TNT. ...................................................... 34
Table 1.3: Characteristic Raman modes of DNT. ..................................................... 35
Table 1.4: Characteristic Raman modes of picric acid. ............................................. 36
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration xvii
List of Abbreviations
AHT 6-Aminohexanethiol
Ag/GNs Graphene nanosheets supported silver nanoparticles
AgNP Silver nanoparticles
AgSMNs Silver nanoparticle-silver molybdate nanowire complex
Au Gold
AuNP Gold nanoparticles
AuNS Gold nanostructures
Au@SiO2 Silica coated gold nanoparticles
BT Butanethiol
DMAB P-dimethylaminobenzaldehyde
DNT 2,4-Dinitrotoluene
eV Electron volt
fM Femtomolar
FRET Fluorescence resonance energy transfer
g/L Gram per liter
H2O2 Hydrogen peroxide
H2SO4 Sulfuric acid
HClO4 Perchloric acid
IMS Ion mobility spectroscopy
KOH Potassium hydroxide
LOD Limit of detection
LQD Limit of quantification
LSV Linear sweep voltammograms
NaOH Sodium hydroxide
M Molar
MF Meso flower shaped nanoparticles
mV Mili Volts
ng/L Nanogram per liter
nM Nanomolar
NO2 Nitrogen dioxide
xviii Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
PA Picric acid
PATP P-aminothiophenol
pAu Gold platform
pAu/AuNS Gold nanostructures developed over flat gold platform
pH Hydrogen potential
pM Picomolar
ppb Parts per billion
QD Quantum dots
RDX 1,3,5-trinitroperhydro-1,3,5-triazine
RSD Relative standard deviation
SAM Self-assembled monolayer
SEM Scanning electron microscopy
SERS Surface enhanced Raman spectroscopy
SiNWs Silver-coated silicon nanowire
SiO2 Silicon dioxide
SPR Surface plasmon resonance
SSEF SERS substrate enhancement factor
TEM Transmission electron microscopy
TNT 2,4,6-Trinitrotoluene
UV-Vis UV-Visible spectroscopy
WAI Wide area illumination
XPS X-Ray photoelectron spectroscopy
4α-CoIITAPc 1,8,15,22-tetraaminophthalocyanatocobalt(II)
Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration xix
Acknowledgements
“Scientific research is one of the most exciting and rewarding of occupations”
-Frederick Sanger.
In the name of Allah, the Most Gracious, the Most Merciful.
Alhamdulillah, all praises to Allah for the strengths and blessings throughout this
challenging journey of completing my research and compiling this thesis. It has been
an overwhelming experience with lots of ups and downs. For that, I would like to
take this opportunity to thank those that have given me never-ending help and
support, of which the success and completion of my work in this thesis would have
been extremely difficult.
If not for my supervisory team, I would not have been able to produce publications
and complete my study within the time frame. Special appreciation goes to my
supervisor, Dr Emad Kiriakous. I am truly grateful for his constant support, advices
and guidance throughout my candidature. His constructive comments and
suggestions throughout the experimental work and thesis writing have been
invaluable. In addition, I appreciate the guidance from my associate supervisor, Adj.
Prof. Peter Fredericks. Thank you to Prof. Godwin Ayoko and Dr Helen Panayiotou
for their timely feedback on improvement of my thesis. I would also like to
acknowledge Arumugam Sivanesan who gave me a big helping hand in my work.
Many thanks for all the help towards the completion of my research.
I am grateful to both University of Malaya (UM) and the Ministry of Higher
Education Malaysia (MOHE) for sponsoring my studies and for their financial
support. I would like to specially thank both parties, as well as QUT, for providing
me the opportunity to pursue this research in which I have gained much experience.
Honourable mention extended to Dr Llew Rintoul for his expertise and advices on
the many aspects of Raman Spectroscopy. My acknowledgement also goes to all the
technical staff for their knowledge, assistance and providing instrument access
xx Raman Spectroscopy Techniques For Detection Of Energetic Materials At High And Low Concentration
during the course of this work. In addition, I sincerely appreciate the help,
cooperation and feedback from my fellow Raman group members (Priyanka, Roland,
Josh and Juanita) and colleagues. I would also like to express my appreciation to
QUT Research Student Centre and QUT Science and Engineering Faculty HDR
Support Team for their support during my candidature.
To my awesome friends, I would like to extend my deepest gratitude to all of you for
your tremendous support, kindness and comforts during this intense and interesting
period. Thanks for constantly keeping me sane throughout this journey. To those
who contributed one way or another your kindness means a lot to me and for that I
am thankful. Thank you very much. For every one of you, I can only pray that God
will bestow goodness upon you.
Last but not least, I dedicate this thesis to my family, especially to my parents. Their
endless prayers and motivation kept me going despite the many difficulties. I am
indebted to them for their unconditional love and unequivocal support.
1
Statement of Social Impact
Explosives are a major cause of organic pollution, with TNT pollution being the
most widespread. The accumulation of explosive residues over the years at war sites,
military training ranges, mining industry and sites of explosive manufacturing have a
serious impact on the environment, and thus threaten the ecosystem and human
health. Most explosive compounds, including TNT, are toxic, mutagenic, and highly
energetic. Bioremediation by plants and bacteria are able to degrade explosive
compounds to a small extent. Nevertheless, TNT is a highly recalcitrant xenobiotic
compound and chemical persistence of TNT at contaminated sites suggests that the
microbial activity in contaminated soils is incompetent in degrading TNT at an
appreciable rate.
This thesis presents the development of a rapid, sensitive and reproducible
spectroscopic method for the detection of TNT in forensic and environmental
applications. Selective recognition of TNT in water and soil were demonstrated with
new aminothiol molecules. Nanoparticles and nanostructured substrates, prepared by
cost effective methods, are utilized as sensitive platforms for the detection of TNT by
surface enhanced Raman spectroscopy. Handheld Raman spectrometers are now
commercially available and can be used in combination with the developed methods
from this thesis for routine and in-field analysis of TNT.
The work presented in this thesis paves the way for effective detection and
monitoring of explosives residues in law enforcement and environmental health
applications. This study will therefore have a positive impact on the safety and
wellbeing of the general community. The presented work is of interdisciplinary
nature and therefore would be of interest to the wider scientific community in
material science, sensor, Raman spectroscopy, environmental monitoring, analytical
chemistry and forensics.
2
Introductory Remarks 3
Introductory Remarks
This thesis titled “Raman spectroscopy techniques for detection of energetic
materials at high and low concentrations” investigates both functionalized gold
nanoparticles and functionalized nanostructured substrates for the sensitive and
selective detection of TNT. Their applicability in forensic and environmental
applications was also explored. Our research focused on Raman spectroscopy,
specifically surface enhanced Raman spectroscopy (SERS) because of the
advancements in this technique down to single molecule sensitivity. The main aim of
this study is to fabricate a simple nanosensor of high sensitivity, selectivity and
reproducibility for SERS analysis. Current SERS based sensors lack reproducibility
and stable Raman enhancements for widespread practical use. This research puts into
context the importance of substrate choice and the selection of an appropriate
recognition molecule to achieve reliable and efficient SERS detection.
As a proof of concept, cysteamine-functionalized gold nanoparticles were utilized to
recognize and detect TNT at trace levels. The well-known Meisenheimer complex
formation of TNT with electron-rich nucleophiles, such as amines, was exploited for
targeting TNT molecules. The required amine moiety combined with the high
affinity of thiols to gold prompted the preference of aminothiols as recognition
molecules. The advantage of cysteamine being a small molecule, which would bring
the target molecule closer to the metal surface, makes it an ideal candidate for
increased SERS signal through the formation of Meisenheimer complex. Shorter
chain lengths exhibit better interaction with the electric field compared to longer
chains. Achieving a low detection limit below the practical requirements of TNT
contamination in wastewater showed the sensitivity and potential application of this
methodology. In addition, the selectivity of Meisenheimer complex formation over
two closely related molecules, 2,4-DNT and PA was established. This was supported
by UV-Vis and Raman studies indicating lack of response in the absence of TNT.
Once the methodology was established, an attempt was made to enhance SERS
activity by tuning the surface properties of SERS active substrates. Colloidal
4 Introductory Remarks
nanoparticles suffer from random aggregation and thus induce uncontrollable hot
spots. This results in generation of irreproducible SERS signal, which is not practical
for routine analysis. By employing the same strategy on a different SERS active
substrate, increase in SERS activity was obtained. Tuning the deposition of a closely
packed layer of gold nanostructures on a flat gold platform increased the abundance
of hot spots present and thus providing an increase in SERS signal. Improved and
reproducible SERS signals were obtained with the usage of nanostructured substrates
in comparison to functionalized nanoparticles. The homogeneity of this
nanostructured substrate improved the relative standard deviation from 8% to 4%
when utilized under wide area illumination. The low detection limit of 100 fM
prompted further expansions to examine environmental samples. 1 nM TNT extracts
from spiked soil sample, without any complex sample pre-treatment, showed
comparable SERS signals with laboratory prepared solutions.
A disadvantage of cysteamine however, is that the majority of the molecules prefer
to orient themselves in a gauche formation on metal surfaces. This impedes TNT
detection by Meisenheimer complex formation, as the amine end of the recognition
molecule is no longer free to bind other than to the gold surface. A different
aminothiol was utilized to expand the versatility of this approach. 6-
aminohexanethiol was elected as a suitable candidate to replace cysteamine because
of its ability to maintain the upright position when assembled on metal surfaces, thus
increasing the chance of TNT molecules to be captured by 6-aminohexanethiol. Long
carbon chains are known to form a dense monolayer on substrates. This may
decrease SERS activity with the overcrowding of 6-aminohexanethiol with less space
for TNT to position itself appropriately. Building upon that knowledge, the potential
of a mixed monolayer of 6-aminohexanethiol with butanethiol, a small molecule with
a non-functional end group, was investigated for a better structured assembled layer.
A detection limit of 100 fM was achieved, with TNT extracted from soil samples
recorded at picomolar levels. The reproducible, sensitive and selective detection
achieved in this study illustrates the potential of the substrate to be used for in-field
detection in a variety of applications.
The flow chart on the next page graphically represents the structure and flow in
which this thesis was compiled.
Introductory Remarks 5
Sensitive and selective detection of TNT using various recognition molecules for SERS analysis
Chapter 2: Cysteamine-
functionalized nanoparticles for
selective SERS detection of TNT in
forensic applications
Chapter 3: Reproducible
nanosensor for SERS detection of
TNT in environmental
applications
Chapter 4: Aminohexanethiol mixed monolayer
nanosensor for SERS detection of
TNT in the environment
Chapter 1: Introduction and Literature Review
Rapid detection of TNT in
aqueous media by selective
label free surface enhanced
Raman Spectroscopy, Talanta,
2015. 134: p 732-738
A homogeneous surface-enhanced
Raman scattering platform for ultra-
trace detection of trinitrotoluene in
the environment, Analytical Methods,
2015. 7(9): p 3863-3868
Molecular recognition of 2,4,6-
trinitrotoluene by 6-
aminohexanethiol and surface-
enhanced Raman scattering
sensor, Sensors and Actuators
B, 2015. 221: p 273-280
6 Introductory Remarks
Introduction 7
Chapter 1: Introduction
“Physical evidence cannot be wrong; it cannot perjure itself; it cannot be wholly
absent. Only its interpretation can err. Only human failure to find it, study and
understand it can diminish its value.” – Paul Kirk
The global increase in usage of energetic materials has increased the demand for
rapid, reliable and sensitive detection of these substances. It has been a constant
struggle to effectively monitor and detect the many different types of explosives all
over the world due to the limitations of available systems. Continuing threats of these
materials is extensive and not limited to criminal investigations but also encompasses
border security, industrial management, environmental safety control as well as the
medical field. With safety and security being the main concern, it is critical to devise
sensors as a first form of defence against these hazardous substances at trace levels.
In addition to the need to identify explosive threats, remediation of contaminated
soils is of interest to governments, as these molecules are commonly used by the
military and are highly toxic to the environment. Therefore, there is an emphasis for
the need of a sensitive and selective method for trace detection of explosives in soil,
groundwater and buried land mines. Current sensing technologies such as
chromatography, neutron activation analysis and ion mobility spectrometry tend to
be either bulky, lack sensitivity, expensive, too sophisticated to handle, have lengthy
analysis times and may require extensive sample preparation or sample alteration [1-
8]. These challenges severely hinder practical real-world applications and
development of new alternative approaches is required. This present research aims to
address this problem with efforts to provide a more rapid, reliable, simple, versatile
detection and identification system.
1.1 BACKGROUND
Explosives encompass a variety of compounds and hence, can be classified in a
variety of ways. One such way is categorization based on their chemical structure
8 Introduction
because of the similarities in their chemical structures. An advantage of this
classification is the indication of how the explosive compound will behave. The
commonly used military explosives contain the nitro group [8]. The five major
chemical categories are nitroalkanes, nitroaromatics, nitrate esters, nitroamines and
peroxides (Table 1.1).
Structure Name Abbreviation Class
2,3-Dimethyl-2,3-
dinitrobutane
DMNB Nitroalkane
4-Nitrophenol NP Nitroaromatic
2,4-Dinitrotoluene 2,4-DNT Nitroaromatic
2,4,6-Trinitrotoluene TNT Nitroaromatic
Picric acid (2,4,6-
trinitrophenol)
PA Nitroaromatic
1,3,5-trinitroperhydro-1,3,5-
triazine
RDX Nitroamine
Tetrahexamine tetranitramine HMX Nitroamine
2,4,6-Trinitrophenyl-N-
methylnitramine
Tetryl Nitroamine
/nitroaromatic
Introduction 9
Structure Name Abbreviation Class
Nitroglycerin NG Nitrate ester
Pentaerythritol tetranitrate PETN Nitrate ester
Triacetone triperoxide TATP Peroxide
Hexamethylene triperoxide
diamine
HMTD Peroxide
Table 1.1: Selected examples of common explosives.
Nitroaromatic explosives are commonly used for military, industrial purposes and
terrorist activities. The presence of one or more nitro (-NO2) functional groups
contributes towards the compounds highly explosive nature. 2,4,6-trinitrotoluene
(TNT) and its degradation product, 2,4-dinitrotoluene (2,4-DNT) are frequently
targeted in explosives detection due to their widespread use as primary components
in improvised explosive devices [9]. These nitroaromatic explosives are electron-
deficient aromatics that can form colored complexes with electron-rich compounds
through nucleophilic aromatic substitutions reactions. In contrast, peroxide
explosives lack both nitro groups and aromatic rings. This group of explosives are
difficult to be detected by most detector systems, which focus on detection of nitro
containing compounds. However, organic peroxides such as triacetone triperoxide
(TATP) and hexamethylene triperoxide diamine (HMTD) are strong oxidants that are
highly sensitive to temperature, friction, and impact and thus can be more powerful
than military analogs. Their structures include free electrons or π-bonds, which
allows screening through electron transfer or oxidation reactions as well as
vibrational spectroscopy [10].
10 Introduction
Detecting trace explosives is challenging due to the intricacies in dealing with ultra-
low amounts of sample [11]. Challenges in the detection of illicit and hazardous
substances include limited sample size and the presence of other substances that may
cause interferences during analysis, swamping the signal of the analyte [12, 13]. The
unpredictability of various complex mixtures used to fabricate improvised explosive
devices increases detection difficulty. The potential presence of unknown substances
introduced by the environment complicates detection further. This fact reiterates the
necessity for a technique that distinguishes between one explosive from another.
Illicit substances are often not present in pure form and are in low abundance, some
down to the picogram level. It should be pointed out that it is difficult for a single
technique to offer complete unique identification of each and every explosive
material. A variety of complications such as sample matrix, substrate matrix,
uncontrollable environmental conditions and vapor pressure may introduce bias into
the test outcomes [11, 14].
1.2 CURRENT TECHNIQUES FOR EXPLOSIVE DETECTION
This section will briefly cover some of the most common explosive detection
techniques currently used, comparing their advantages and disadvantages. Various
analytical methods exist for the molecular detection of explosives. This includes ion-
mobility spectroscopy (IMS), chromatography, optical sensors, molecularly
imprinted polymers, trained canine teams, terahertz spectroscopy (THz) and X-ray
dispersion [3, 5, 6, 15, 16].
1.2.1 Canine detection
The use of canines is a well-established methodology for explosives detection as
research has shown that the canine olfactory system has a larger surface area than a
human, allowing higher sensitivity to trace vapour [17]. Well-trained canines can
easily discriminate between target molecules, and can be deployed easily to detection
sites. Although a versatile and reliable real-time detector, canines are expensive to
train and tire easily, making widespread use challenging [3, 5, 10]. Detection using
technology, on the other hand, can be fully utilized around the clock. The limited
ability of determining detection levels also hampers the effectiveness for quantitative
Introduction 11
detection. The capabilities of other animals such as rats, pigs and bees are also being
assessed towards detection of illicit substances [18-20]. The success of animals for
fieldwork in identification of explosives has brought about advances in olfactory type
sensors to imitate explosive sniffing animals. Currently, artificial olfactory type
sensors, known as electronic noses, are being developed to overcome the issues
associated with using animals especially canines [17, 21, 22].
1.2.2 Ion Mobility Spectroscopy (IMS)
Ion mobility spectroscopy (IMS) is currently used commercially for detection of
hazardous materials and explosives particularly at airports and high security areas.
IMS analysis detects the molecular mass of vapour or swipe sample and positive
identification requires a comparative analysis with a library of known standards [23].
The underlying principle of this technique is the ionization of sample materials,
which drifts through the instrument based on its mass/charge ratio due to the applied
electric field. The different mass, charge, shape and size of different ions influence
the migration time of ionized molecules. Each ion mobility band produced
corresponds to each ion species and thus presents a fingerprint spectrum of the parent
compound [3, 23].
Sensitive detection can be achieved for clean samples and when targeting only a few
explosive compounds. Development of current hand held IMS devices has provided
a rapid field-deployable detection system. The major disadvantage, which limits
selective and sensitive detection, is the lack of positive identification and instability
in the presence of interfering matrix molecules or contaminants [24]. This
technology does not appeal for a broader use as it suffers from the need of skilled
personnel for instrument calibration.
1.2.3 X-ray scattering
X-ray based systems are widely used in airports, whereby objects are scanned on
both sides. Backscattering X-ray screening utilizes high-energy X-rays that can
penetrate through luggage and clothing. This effective screening process allows
detection with high-resolution images of concealed items based on shape
visualization. The weakness of X-ray based techniques is its incapability to provide
12 Introduction
quantitative data and thus data obtained does not incorporate material identification.
This screening methodology is also not portable and has not achieved trace level
detection [3, 10].
1.2.4 Colorimetric assays
Colorimetric explosives sensors rely on a selective chemical reaction between the
explosive target molecule and an indicator molecule, leading to an observable change
in color in the presence of the target molecule. Meisenheimer or Griess reactions are
widely used to create a new chromophore from the reaction between the explosives
residue and indicator molecule [15]. The formed chromophore produces a visible
color change, indicative of the presence of explosive. Colorimetric sensors can be
used as portable sensors due to rapid detection, simple sample preparation
procedures, and ease of interpretation [3]. However, substances with similar
characteristics are difficult to distinguish with this method, as the indicator molecule
is typically a broad detector of a certain characteristic, group or class. This is one of
the reasons why colorimetric kits available for field detection of explosives are
mostly used as a presumptive test before being sent for further confirmatory analysis.
Furthermore, certain colorimetric assays such as the Greiss reaction require multiple
steps performed sequentially to form an azo dye for determination of the presence of
the nitrite anions [10].
1.2.5 Fluorescence assays
Fluorescence detection can provide good sensitivity towards the screening of
explosive analytes. The ability for fluorescence emission to be measured on a low
background has made it a well-established technique. Nevertheless it is limited to the
visible range excitation wavelength. Fluorophores used for explosives sensing
include both organic and inorganic molecules such as polymers, thin films, sensor
arrays, and fiber optics. The reliance of fluorescence sensors on excited-state charge-
transfer for detection permits a wider scope for the chemical composition of the
indicator and analyte. Much research has focused on fluorescence quenching, in
which the presence of the target explosive reduces the intensity of light emitted by
the fluorescent indicator [7, 10, 25]. This is the inverse of colorimetric assays.
However, fluorescence methods are neither of high selectivity nor specificity and do
Introduction 13
not give rich information about the identity or the chemical fingerprint of the
detected molecule. Therefore, they may be subject to false positives. Other
limitations with fluorescence sensors include photodegradation and photobleaching
of fluorescent materials [10].
1.2.6 Tetrahertz spectroscopy (THz)
This emerging technology utilizes far-infrared radiation in the tetrahertz (THz)
frequency (0.1-10 THz). Being a spectroscopy technique, distinguishing between
samples is based on the different characteristic spectral features produced from
differential absorption of the sample under THz radiations. Similar to X-ray
scattering, THz imaging is able to detect concealed items through packaging. Its fast
processing time and high resolution is challenged by its expensive cost and power
requirements, which hampers it from being field deployable for routine analysis on
site. Another disadvantage is that THz spectra quality is affected in the presence of
water. Analysis of solution-based samples is not suitable with this technology. This
technique is still in its infancy and its full potential is still being explored [26, 27].
The disadvantages portrayed by these methods demonstrate that none are ideal albeit
the advantages they provide. A significant deficiency is the susceptibility to false
positive or false negative due to environmental contaminants and interfering
compounds which can mask the target sample [3, 6, 28, 29]. Significant detrimental
effects will result from both false positive or false negative [30]. Some of these
approaches either cause sample destruction or isolation of sample, both of which do
not allow preservation of the original sample [5, 6]. These shortcomings do not
favour low quantity samples.
1.3 RAMAN SPECTROSCOPY
Among the most effective techniques for explosive detection are the spectroscopic
methods (IR, UV, and Raman). These techniques are advantageous because they
provide sensitivity, cost-effective instrumentation, portability, and maintain the
ability to detect a wide range of chemical explosive classes. The poor specificity and
high water absorption in infrared spectroscopy (IR) limit its broader use for the
14 Introduction
detection of explosives [30]. In contrast, Raman spectroscopy is more promising for
the detection of explosives with its ability to identify characteristic vibrational
features.
Raman spectroscopy for the detection and identification of illicit substances has
provided new opportunities because of its advantages being non-invasive and the
ability to process various types of samples whether in liquid, solid or gas state [2, 6,
13, 31]. This spectroscopic method provides a unique molecular fingerprint spectrum
of the sample even though it requires only a small sample volume with minimal or no
complex sample pre-treatment. The little-to-no water Raman signal provides an
added advantage for analysis of aqueous samples without worry over competing
Raman signals from water molecules [32].
The Raman mechanism lies in the changes of the molecules vibrational state when
illuminated by light [6, 33]. A Raman active molecule is one where the vibrations in
the molecule lead to a change in polarizability. In general, light can be either
absorbed or scattered when it interacts with matter. Figure 1.1 depicts the schematic
energy level diagram for various light interaction mechanisms. Raman spectroscopy
deals with light scattering. Scattering occurs in the presence and absence of available
suitable pair of energy levels to absorb the radiation during the interaction between
light and the molecule. The output frequencies of scattered light from a molecule are
either elastic Rayleigh scattering or inelastic Raman scattering. When light that
interacts with the sample is scattered in the same frequency as the incident light with
no energy loss, it is known as Rayleigh scattering, whereas change in frequency of
the incident light leads to inelastic light scattering called Raman scattering. If the
inelastic light gives up its energy to the sample and shifts towards a lower frequency
it becomes Stokes scattering. Shift towards a higher frequency than the incident light,
on the other hand, leads to a blue-shifted frequency referred to as anti-Stokes
scattering [2]. Anti-stokes scattering involves the excitation of molecules that are
initially present at higher vibrational states. Upon excitation, the molecules tend to
relax to a lower vibrational state, thus emitting photons of relatively shorter
wavelengths.
Introduction 15
Figure 1.1: Energy level diagram of the various light scattering processes.
This difference in frequencies between the incident light and inelastic scattered light
corresponds to the vibrational modes of the target molecule [34]. The molecular
vibrational modes in turn are dependent on its composition and the different
functional groups in a molecule give rise to characteristic shifts in frequency [33].
Hence, every molecule would have a characteristic Raman spectrum arising from the
Raman shift against the intensity of the scattered photons. This molecular fingerprint
conveying chemical and structural information based on intensities and peak
positions makes Raman spectroscopy an attractive analytical technique [34, 35]. Its
high chemical specificity facilitates not only identification but also differentiation of
highly similar molecules or the different molecular orientations of a molecule.
Nevertheless, Raman scattering suffers setbacks due to its low intensity inelastic
scattering as the incident light are mostly scattered elastically without frequency
change. Only a miniscule amount of incident light experience Stokes and anti-Stokes
scattering, giving it poor sensitivity. Even fluorescence emissions are at higher
magnitudes than the Raman effect causing interference and restrict Raman
measurements [2]. The weak Raman scattering effect leads to its inability to detect
trace contaminants especially in real life applications. The low sensitivity drawback
of Raman spectroscopy has been overcome with the emergence of surface enhanced
16 Introduction
Raman spectroscopy (SERS), which is known to be sensitive down to single
molecule level.
1.4 SURFACE ENHANCED RAMAN SPECTROSCOPY (SERS)
SERS is a powerful spectroscopy technique that offers the means to greatly enhance
the weak spontaneous Raman signals by orders of magnitude [31]. The ability of
SERS to produce massive enhanced signals evidently benefits trace level detection.
The amplification of the Raman signals occurs by placing the molecules within close
vicinity to noble metal surfaces such as gold and silver [12, 31]. As a result,
optimization of the metal surface is vital to achieve an effective sensor. Fleischmann
et al. [36] initially observed this phenomenon through experimental discovery of
strong Raman signals when pyridine was absorbed on a roughened silver electrode. It
was proposed that the roughening process increased the surface area thus allowing
more pyridine molecules to be absorbed, enhancing the Raman scattering efficiency.
Independent studies by Van Duyne [37] and Albrecht et al. [38] brought about the
findings that electric field enhancements and molecular interactions with the metal
surface are responsible for Raman scattering respectively. The capability of SERS to
detect low amounts of substance within a small area of the focused laser makes
SERS well suited for trace detection. The attractive combination of sensitivity of
SERS with the instrumental benefits of Raman spectroscopy enables highly specific
molecular information on the microscopic scale [13]. The high sensitivity of SERS
has shown sensitivity down to single molecule detection [39-41].
The success of SERS relies on advances in engineering the features of SERS
substrates, in order to enhance the electric field around the metal surface.
Understanding the role of the major parameters affecting the SERS intensity would
lead to better control of fabrication of efficient SERS active substrates. The drastic
enhancement in Raman signal is understood to have come from the contribution of
two mechanisms: electromagnetic enhancement and charge transfer or chemical
enhancement [31, 33]. The electromagnetic enhancement, which is the major
contributor, results from the excitation of the localized surface plasmon resonance
(LSPR) (Figure 1.2) [42-45]. On the other hand, the chemical enhancement
Introduction 17
mechanism is a result of the interaction between the electrons in a molecule with the
metal surface.
Figure 1.2: Schematic representation of the surface plasmon electron oscillation.
Adapted from reference [45].
Molecular polarization is fundamental to Raman scattering. The strength of the
induced dipole moment is related to the interaction of the electron cloud with the
electric field. The strength of the induced polarization is represented by the following
equation [46]:
μ = αE Equation 1
Where:
α = molecular polarizability constant
E= magnitude of the incident electric field
1.4.1 Electromagnetic enhancement
Electromagnetic enhancement focuses on the E term in Equation 1. The interaction
between the electric field component of the incident light and metal nanostructure
18 Introduction
induces free electron oscillations of the metal, creating a dipole around the particle,
resulting in LSPR [47, 48]. The LSPR strength is dependent on the interaction of
both incident and scattered light with the metal substrates. For the plasmons to
remain localized, the wavelength of the incident light has to be larger than the
nanoparticles. Large electromagnetic enhancements are created when the correct
wavelength of incident light is resonant with the collective oscillation of the metal
electron conduction [43]. In essence, the induced dipole in the absorbed molecule
amplifies the Raman scattering. Excitation of the LSPR generates enhanced field
intensities at the surface of the roughness feature of metal nanostructures. The
Raman signal enhancement by the large electromagnetic fields on the small gaps
between metal substrates, known as hot spots, allows SERS to detect trace levels of
target substances in the screened sample [49]. Thus, robust nanostructures with gaps
or crevices that enhances the electromagnetic field are ideal for SERS [50].
1.4.2 Chemical enhancement
The chemical enhancement mechanism, also known as charge transfer is dependent
on the chemical nature of the absorbed substance and its interaction with the metal
nanostructure. This charge transfer transition is analyte dependent, whereby the
polarizability change that leads to an enhanced Raman signal relies upon the
absorption process [51]. The increase in distance between analyte and metal surface
significantly decreases the SERS intensity acquired [52]. This electron transfer
complex formation between the molecule and the metal is considered to be more
complex and contribute less to the overall SERS enhancement. Only certain
molecules will undergo this effect, as it requires the formation of a chemical bond
with the surface for an enhancement to occur. Although SERS is largely dependent
on the electromagnetic enhancement, an understanding of the chemical
enhancements allows interpretation of changes in relative Raman intensities [53].
1.5 SERS SUBSTRATES
In the field of SERS, one of the major challenges is the fabrication of efficient SERS
substrates with uniformity over the whole substrate surface. Optimum SERS effect is
strongly dependent upon the surface morphology of the substrate. Tailoring the
Introduction 19
nanoscale features of SERS substrates affects both the LSPR and molecular
absorption of target molecules. As mentioned in section 1.4.1, ideal SERS substrates
must be able to support LSPRs for enhanced signals. Significant efforts have been
made for fabrication of efficient SERS substrates. To date, numerous designs of
SERS surfaces such as roughened electrodes, metal nanoparticle films and metallic
nanostructures have been suggested [54]. The metal employed as active SERS
substrates is an essential factor for the sensitivity and intensity of the Raman bands
[33]. The most frequently used SERS substrates are gold and silver due to their
useful properties and strong absorption characteristics. Much attention in both early
and recent studies is towards colloid-based SERS since colloids are relatively easy to
prepare and have the tendency to produce large enhancement factors. Typically,
metallic colloids are prepared through chemical reduction reaction in solution.
Common metallic colloids include citrate-reduced colloids and borohydride-reduced
colloids, in which the metal is both reduced and stabilized by sodium citrate and
sodium borohydride respectively [33]. The amplification of Raman signal is
attributed to colloid aggregation that leads to the formation of hot spots [54]. Gaps
between close proximity molecules enhance SERS signals by up to factors of 108
[31]. The size, shape and degree of aggregation are some factors that greatly
influence the useful properties of nanoparticles [43, 48, 55, 56]. Nevertheless,
metallic colloidal particles are challenging to fabricate reproducibly and not
consistently effective for routine analysis. Variability between different measurement
spots or batches of colloids would cause large fluctuations in SERS enhancements
and thus produces irreproducible SERS spectra [32].
The large variations in SERS signals have resulted in advancements in development
of SERS substrates to a more regularly arranged and highly ordered platform [57].
Fabrication and optimization of SERS substrates is a current active field of research
in nanotechnology. Expansion of metal particles in many different forms from
nanoparticle shape varieties to tunable SERS substrates are scattered through
literature with the emphasis on homogeneity, reproducibility and better control over
their properties [16, 43, 49]. Fabrication of nanostructured substrates is one example
of a fast and simple nanofabrication. Nanofabrication on the metal substrate itself is
more straightforward than chemical synthesis of nanoparticles in solution. The
20 Introduction
advantage of this methodology is the ability to have a better control on the size,
shape, stability and uniformity of the SERS substrate [58]. Controlling these
parameters is crucial for an efficient SERS response as the LSPR are influenced by
these properties. Electrochemically roughened gold or silver surfaces not only
support surface plasmon resonance but also combines both SERS and
electrochemistry for an added advantage [59]. Implementing SERS in analytical
chemistry requires sensitivity, uniformity, stability and reproducibility to increase the
potential usefulness of the technique for practical applications [14, 33]. This
motivated us to exploit the benefits of SERS-active nanostructured substrates and
SERS phenomenon for improved trace explosive detection.
The design of SERS-active substrates is an important aspect for further exploitation
of SERS applications. There is no universal best SERS substrate but the metal used,
the design of the substrates and the substrate-molecule interaction are fundamental
parameters that need to be considered for substantial enhancement of the Raman
signal [13]. In general gold has been of widespread use for its stability, being less
susceptible to oxidation, and greater control over particle size [32, 33].
1.5.1 Gold substrates
The type of substrate compatible for a certain application is dependent on the
properties required for the application itself. Gold may not be superior to other
metals such as silver or palladium for certain applications. Nevertheless, gold was
chosen as a suitable substrate for this study based on the following characteristics
[32, 60]:
1. It is easy to synthesize gold colloids;
2. Gold possess tunable optical properties;
3. Common substrate utilized in existing spectroscopic and analytical
techniques;
4. Gold is chemically inert; it does not oxidize readily in air at room temperature
and does not react with atmospheric oxygen or most chemicals;
5. Has a high affinity for thiols. Self –assembled monolayers (SAMs) of thiols
on gold are stable for weeks;
Introduction 21
6. Support LSPR in the visible to near-infrared (NIR) regions;
7. Offers a platform for easy surface functionalization.
These characteristics make gold a robust platform for chemical detection of target
analytes by manipulating its morphological features to obtain efficient SERS
response. Although silver is widely studied and seen as a high-quality substrate for
SAMs, it is susceptible to oxidation [61]. This shortcoming is not a practical
alternative for handling samples under atmospheric conditions and especially
problematic in field. Other materials such as copper, liquid mercury and palladium
do offer some useful characteristics. However, studies of SAMs formation on these
materials are still in infancy. This gives an additional advantage of gold over other
materials because the absorption and reaction of molecules onto the active surface is
better understood [33]. The mentioned advantages of gold are crucial aspects for the
preparation of stable and highly active SERS substrates to support the applicability
for trace chemical analysis.
1.6 SELF-ASSEMBLED MONOLAYERS (SAMS)
The possibility of molecular recognition at very low concentrations in the presence
of metal nanomaterial surfaces allows SERS to be one of the most powerful and
versatile analytical tools. In addition to sensitivity, another important feature is
specificity, which can generally be attained by controlling the chemical moiety
around the metal surface [4]. Despite the fact that SERS offers a sensitive, easily
miniaturized system, it suffers from problems that may limit its applications for trace
detection. Bare substrates are generally sensitive to any Raman-active analyte that
interacts with the metallic surface of the nanoparticle [44]. This is one of the major
issues that need to be addressed in order to distinguish the target substance from
matrix effects and background clutter, particularly in complex samples [14, 57].
Competing moieties with higher affinities, especially in environmental samples
containing multiple species, could obscure the signal from the target substance.
Although each component in the mixture would present its own unique SERS bands,
it would be challenging to separate each individual component from the complex
spectra [44]. As most energetic materials contain the characteristic –NO2 group,
22 Introduction
interference from similar structured molecules can also mask the Raman signal of the
target sample and may produce false alarms [16].
Chemical design of SERS surfaces is paramount for application in sample
monitoring under environmental conditions. The assembly of functional groups of
special affinity towards the target molecule, onto the metal surface can improve the
selectivity of the SERS detection of the target molecule significantly [57]. This leads
to a more robust detection mechanism of the target molecule in complex matrices.
Self-assembled monolayers (SAMs) are organic assemblies of molecular
constituents, which provide a simple and flexible system to tailor desirable qualities
of the metal surface [60]. SAMs absorb spontaneously on metal surfaces through
their head-groups, which have favourable affinity towards the metal and a terminal
functional group on its other end to control surface properties. A wide variety of
ligands such as hydroxyl, carboxylate, thiols and amines can form SAMs on gold.
However, thiols with sulphur atoms as head-groups are generally used for surface
attachment as they have a strong preferential absorption to gold surfaces [62].
Various SERS studies have taken advantage of employing surface functionalization
of a recognition molecule via the use of thiol compounds such as aminothiophenol,
L-cysteine, 4-mercaptobenzoic acid, N-acetyl-L-cysteine, and L-cysteine methyl
ester hydrochloride for its strong affinity for metal surfaces through click chemistry
[4, 63-65].
As mentioned earlier, the tail groups that point outwards from the metal surface
provides the functionality of SAMs. Modification of the metal surface induces an
improved interaction between analyte and substrate, providing the ability to target
specific molecules and prevent cross-reactivity from interfering molecules. This
would consequently enhance the selectivity towards the target molecule. Many of
these detection schemes rely on the aggregation of the functionalized nanoparticles
for presence of the target molecule [15, 64, 66-68]. Hexaazamacrocyclic ligand,
cysteamine, 1,2-ethylenediamine, aminothiophenol and curcumin are some examples
which have been used to modify the surface of gold nanoparticles for sensitive
detection of TNT based on changes in aggregation, visual colorimetric and
absorption spectra [15, 64, 66-68]. These types of assays are rapid and simple but
they are however prone to unspecific aggregation from a variety of other molecules
Introduction 23
that easily aggregate nanoparticles, which in turn can generate erroneous results. As
an alternative, more robust strategies are required to tailor the functionalized
particles to produce a fingerprint of the target analyte bound. As a result, SERS-
based sensors for explosives and illicit substances will be highly valuable. Even
though SERS signals do not require direct contact between metal surfaces and the
adsorbate, the electromagnetic enhancement decreases with the increase of
interaction distance. Kneipp et al. [31] demonstrated that the SERS intensity is
proportional to the fourth power of the excitation frequency, showing the importance
of the coupling distance between the metal and the molecule.
To accomplish the objective of trace detection by SERS analysis, the affinity of a
metal surface can be increased by its functionalization with recognition molecules
that can selectively interact with the target molecule. Several key features for an
effective recognition molecule are:
1) A strong affinity of the head functional group for the metal surface and the
terminal functional group for the target molecules;
2) High selectivity towards the target analyte;
3) Lack of band overlap with the Raman features of the target analyte;
4) Ability to stimulate formation of hot spots, which would lead to enormous
electromagnetic intensification;
5) Be of small size to bridge the target analyte close to the metal surface.
Other than increasing the affinity towards specific analytes, SAMs function as a
substrate stabilizer and helps improve sampling precision across the substrate [57].
The SERS spectrum is dependent on the orientation and binding properties of the
molecules at the metal surface. The extent of the interaction between the metal
substrate surface and the target molecule affects the peak shape, peak shifts,
appearance of new bands and peak intensity. Differences in vibrational freedom and
symmetry of absorption would give rise to different peaks being enhanced [69]. The
metal-molecule interaction is somewhat complex, with many factors affecting the
result outcome. Apart from the metal-molecule linkage, metal roughness, size and
shape as well as the space between surface protrusions are some of the features that
play an important role in signal enhancement [33].
24 Introduction
1.6.1 Preparation of SAMs
A widely used procedure for absorption of SAMs on metal substrates is by
immersion of the metal into ethanolic solutions of thiol overnight for about 12-18
hours at room temperature. The long hours are required for optimization of the
SAMs by maximizing the density of the adsorbates. The formation kinetics and
structure of the resulting SAM depends on a number of factors, which are
summarized as follows [70-74]:
Choice of solvent. The preferred solvent for preparation of SAMs is ethanol. Ethanol
is widely used because of its suitability for solvating alkanethiols of varying chain
lengths and polarity.
Concentration and immersion time. The adsorption process undergoes two phases:
a rapid adsorption within the first few minutes followed by a slower process lasting
several hours for complete monolayer formation. Longer immersion times are
required for lower concentration thiols. As mentioned earlier, the typical immersion
time when exposed to thiols above 1 mM is between 12 to 18 hours.
Cleanliness of substrate. Immersion of substrates into thiol solution should take
place within 1 hour of substrate preparation to avoid adsorption of other foreign
materials when exposed to ambient conditions. Minimizing the time between
preparation of substrate and thiol immersion increases the reproducible properties of
SAMs. In the case of prolonged exposure, the substrates can be cleaned with piranha
solution (H2SO4:H2O2) or other strong oxidizing chemicals to displace other
adventitious materials that have absorbed onto the substrate surface.
Temperature. The kinetics of SAMs formation increases with elevated temperatures
above 25°C. The temperature effect is most relevant during the first few minutes of
SAM formation.
Purity of thiols. Although trace amounts of impurities present in thiols solution may
not impede SAM formation, these materials may alter the physical properties of
SAMs.
Understanding these fundamental experimental conditions allows better control of
SAM assembly as it minimizes the amount of defects during SAM formation. Some
Introduction 25
of these factors have a minor effect on SAM formation but effectively controlling
these conditions can lead to a more reproducible assembly of high order.
1.7 ALKANETHIOLS SAMS
The advantage of alkanethiols over other molecules is its thermal stability in air and
in liquid [75]. Alkanethiol monolayers are convenient and versatile as the functional
end group allows electron donor or acceptors to anchor on, introducing various
functionalities onto the metal surface. The varying order of the hydrocarbon tail acts
as a physical barrier for interfering molecules and provides varying thickness of
SAM formation. Longer alkanethiol chains adopt an energetically higher degree of
orientation due to the increase in molecular packing density compared to shorter
chains [57]. A representation of alkanethiol SAM formation is illustrated in Figure
1.3. However, the size of the functional groups may disturb the ordered formation of
SAMs, as a bigger functional group would require more space. Greater tilting would
occur from alkanethiols with bigger sized functional groups [60]. Introducing mixed
monolayer assemblies helps to enhance the high structural order of SAMs.
Figure 1.3: Schematic representation of SAMs formed from alkanethiols on a gold
substrate surface. Yellow circles represent chemisorbed thiol head-groups whereas
the green circles represent the terminal groups of any chemical functionality.
Adapted from reference [60].
26 Introduction
1.7.1 Mixed SAMs
Improved chemical selectivity can be achieved by combining different alkanethiols
on the same platform. Mixed SAMs is simply coadsorption of solutions containing a
mixture of thiols for a more controlled assembly [75]. The dense packing of a single
alkanethiol monolayer creates a steric effect due to close proximity of the alkane
chains. A mixture of two or more ligands to the metal surface can be attached in
well-defined ratios. Enriching the surface with varying compositions of mixed
alkanethiols with different chain lengths and different tail group favours a more
stable packing of the hydrocarbon chains, allowing placement of species at suitable
distances for full functionality [76]. An example of a mixed SAM formation on gold
is depicted in Figure 1.4.
Figure 1.4: Schematic illustration of the possible immobilization of TNT on mixed
SAMs on a gold surface.
Introduction 27
1.8 MEISENHEIMER COMPLEX
Despite encouraging developments in the practicability of SERS for explosive
detection, routine trace analysis require a more stable, reproducible and easy to
fabricate substrate which would allow for a cost effective system. One of the most
widespread energetic materials, which are also known to be a serious pollutant, is
2,4,6-trinitrotoluene (TNT). Efforts to successfully detect trace amounts of TNT with
selectivity are the common focus of this research.
For the sensitive and selective detection of nitro aromatics by SERS, the
development of Meisenheimer complexes have been considered. As shown in Figure
1.5, the Meisenheimer complex is a common intermediate structure formed during a
nucleophilic aromatic substitution reaction [77, 78]. Many TNT sensors being
studied have taken advantage of the interaction of TNT with amine compounds,
which is known to generate Meisenheimer complexes [1, 7, 15, 64, 66, 79-84]. As
nitroaromatic molecules are electron-deficient, they represent ideal targets for
nucleophilic aromatic substitutions. Meisenheimer complexes essentially arise from
the strong donor-acceptor interaction between electron-deficient aromatic rings and
electron-rich amines [4, 81, 84]. The three electron withdrawing –NO2 groups of
TNT make it susceptible to nucleophilic attack by weak nucleophiles such as primary
amines [81, 85-88]. As a result, this complementary pairing can be tailored to enable
selective TNT detection.
Figure 1.5: Nucleophilic reaction between electron-deficient nitroaromatic
compound and an amine (electron-rich nucleophile) to form a Meisenheimer
complex.
28 Introduction
Formation of Meisenheimer complex in the form of colored adducts presents a
positive result and the intermediate structure is resonance-stabilized under favorable
conditions [78]. Different substituents on the aromatic ring cause nitroaromatic
compounds to form Meisenheimer complexes of different colors when reacted with
electron-rich compounds [4]. This leads to a fair degree of discrimination between
the various nitroaromatic compounds.
Amine thiol compounds are perfect nucleophiles that can easily form Meisenheimer
complexes with nitroaromatic explosives as well as attach to the metal surface of
nanoparticles through click chemistry. The –SH group of aminothiol compounds can
form a strong Au-S bond with gold nanoparticles whereas the –NH2 group would
react with the electron deficient nitroaromatic molecules [81].
1.9 SERS DETECTION OF TNT
SERS studies for trace analysis of energetic materials have shown promise in recent
years. Various sensors based on bare nanoparticles and functionalized nanoparticles
have been reported for ultra-trace detection of TNT by using SERS. Jerez-Rozo et al.
[89] pointed out the advantages of TNT detection via alkaline hydrolysis. Both pH
and molecule orientation are two essential considerations for SERS detection of
TNT. Higher signal enhancements were achieved under basic conditions with an
optimal pH of 13. The authors also noted that increase in pH resulted in changes in
molecular absorption geometry of TNT on the metal surface.
One of the earlier works by Kneipp et al. [90] demonstrated a simple SERS detection
of TNT on both colloidal gold and silver. Observations from this study showed that
TNT absorption favours colloidal gold over silver, resulting in higher enhancements
of TNT on gold. The importance of the absorption geometry of TNT onto the metal
surface was also highlighted in this study. The symmetry of the molecule onto the
metal surface determines which SERS bands are selectively enhanced. When TNT
molecules lay flat, the out-of-plane bands will be enhanced, whereas if TNT orients
perpendicular to the surface, the ring-breathing mode is enhanced. Understanding
these essential properties may provide a better strategy for higher sensitivity of TNT
detection. Glembocki et al. [69] experienced drawback whilst utilizing cysteamine-
Introduction 29
functionalized gold and silver substrates to detect TNT. Weak SERS signal was
attributed to the two differing configurations in which cysteamine would adopt on
the metal surface. A higher proportion of cysteamine chemisorbed onto metal surface
in a gauche formation because of its stability compared to the trans conformation
[91, 92]. The tendency for cysteamine to bend back on itself towards a gauche
configuration prevents the interaction of cysteamine with TNT and thus the loss of
signal. The key parameter that needs careful optimization is to ensure the chemical
moiety binding to the metal surface is absorbed in the trans or upright position to
successfully detect the target molecule.
Recent work by Guo et al. [1] resulted in detection of TNT down to 1 pM through
the formation of nano dumbbell structures. The nano dumbbell structures required a
multi-step formation of first assembling the Meisenheimer complex onto the gold
substrates. Positively charged nanoparticles with the attachment of the Raman probe
4-mercaptopyridine were assembled onto the prepared metal substrate before
addition of another layer of negatively charged nanoparticles. The work was useful
for sensitive detection of TNT but the complex procedures for sample preparation is
a big disadvantage for real time detection.
A few groups have reported the detection of TNT down to low levels using the
formation of Meisenheimer complex between cysteine and TNT [4, 81]. These
groups demonstrated the selectivity of their sensor towards TNT over DNT and
nitrophenol where no Raman signal in the presence of DNT and nitrophenol was
observed. The applicability of SERS in realistic conditions however, was not
addressed in these studies. A drawback of utilizing cysteine that cannot be ruled out
is the possibility of the carboxyl group of cysteine to react with the gold surface and
also form a strong hydrogen bond with the amine group of an adjacent cysteine
molecule. The amine group is also able to form salt bridges with carboxylate groups
[93, 94]. Besides that, studies have shown that the free amine and carboxyl group of
cysteine can form complexes with Cu2+ ions [95, 96]. Although cysteine achieved
good SERS sensitivity down to picomolar, this shortfall may deter the full capability
of cysteine to capture TNT molecules especially in real world samples, which are
complex. Thus, a recognition molecule with a single terminal functional group would
be better.
30 Introduction
Hybrid nanomaterials such as graphene nanosheets supported silver nanoparticles
(Ag/GNs) have also been used as a SERS platform for TNT detection. Liu and Chen
[97] achieved a sensitive detection limit of 10 pM by pairing p-aminothiophenol
(PATP) to the graphene-based SERS platform. The interaction between the aromatic
ring of TNT and an amine group formed the Meisenheimer complex to induce the
Raman signals. Despite this advantage, similar Raman results were obtained in the
presence of other nitroaromatic compound such as nitrobenzene, nitrotoluene, DNT
and nitrophenol. The modified PATP-modified Ag/GNs substrate exhibited two
characteristic main peaks, namely, NO2 symmetric stretching vibration (1354 cm-1)
and C=C aromatic stretching vibration (1610 cm-1) not only in the presence of TNT
but with other nitroaromatic compounds as well. Although the Raman intensity for
these bands were not as intense as in the presence of TNT, the superimposed bands
have the potential of masking the TNT signal particularly when existent at a higher
concentration than TNT. This study shows the importance of utilizing a suitable
molecular recognition moiety capable of selecting the target molecule over closely
related molecules to avoid false positives.
Another silver nanoparticle based strategy looks at imprinting molecular recognition
for detection of TNT. Yang et al. [98] coupled PATP with silver nanoparticles-silver
molybdite nanowire complex (AgSMNs complex) to produce p-
dimethylaminobenzaldehyde (DMAB) modified AgNPs arrays. The catalytic
reaction between PATP and AgSMNs complex generated DMAB-modified
AgSMNs complex. The π-donor-acceptor interaction between TNT and DMAB-
modified AgSMNs complex aggregated the nanoparticles, providing hot spots for
efficient SERS enhancement. Their approach successfully achieved a low detection
limit down to 1 pM. Its long-term stability and recyclability due to the generation of
new Ag nanoparticles under laser irradiation, gives this approach an added
advantage. Similar to the work carried out by Liu and Chen [97], this study also
utilized PATP as the recognition molecule to capture TNT. Again, PATP was unable
to deliver high selectivity with similar key spectral features recorded between TNT
and DNT. The DMAB-modified AgSMNs complex was not tested with
environmental samples to test the practicality of this methodology in real world
conditions. Thus, there is a possibility of other interfering molecules in complex
mixture samples, which can swamp the TNT signal. It cannot be stressed enough that
Introduction 31
selectivity plays an important role for ideal trace detection. Holthoff et al. [99] have
reported another molecular imprinting protocol for detection of TNT. In contrast to
PATP, a hybrid molecular imprinted polymer (MIP) consisting of three different
silanes were utilized to achieve non-covalent interactions with TNT. However, the
detection limit achieved with this method is high (3 μM), which limits its efficiency
for real applications.
Fabrication of different shaped gold nanoparticles has also resulted in a highly
sensitive SERS sensor for detection of TNT. Demeritte et al. [80] have synthesized
popcorn-shaped gold nanoparticle-modified single wall carbon nanotubes hybrid thin
film through a multi-step process. Their approach achieved highly sensitive detection
down to 100 fM with selectivity over DNT and 1,3,5-trinitroperhydro-1,3,5-triazine
(RDX). Mathew et al. [100] detected sub zeptomole concentrations of TNT by
utilizing meso flower nanoparticles. Zeptomolar levels are at the single molecule
level and therefore reproducible sensing would require a homogenous SERS
substrate to efficiently pinpoint those single hot spots. Nanoparticles, which were the
basis of both these different shaped nanoparticles, generally suffer from spot-to-spot
and substrate-to-substrate reproducibility. With the known standing of nanoparticles
as inconsistent SERS substrates, there would be difficulty in searching for hot spots
for reproducible detection at very low levels, rendering this methodology not
practical for routine analysis. The difficulty faced by colloid-based SERS substrates
is the production of controlled particle aggregation for a uniform distribution of hot
spots [57]. Hence, a more uniform or homogenous substrate surface is essential for
effective routine SERS analysis.
In spite of progress in current technologies, effective and efficient detection of
explosive compounds remain a challenging task. Indeed, each of these previous
works has their own advantages, demonstrating that the SERS technique has the
potential to achieve low detection limits needed for trace detection. Nevertheless,
their common feature of either presenting a high degree of complexity, lack of
selectivity or not being applied to real samples hinders practical real-time detection.
There is a need to balance between both sensitivity and selectivity to achieve reliable
detection. Based on the challenges posed by explosive materials that have been
highlighted in the preceding sections, it is crucial for the nanosensor to satisfy
32 Introduction
increased sensitivity and selectivity with high accuracy. The analytical platform
should be portable and tolerant to real life environmental conditions to offer potential
paradigm for trace explosive detection.
1.10 RESEARCH OBJECTIVES
Research related to detection of energetic materials utilizing surface-enhanced
Raman spectroscopy has grown in the past decade alongside the ever-expanding
scope of SERS applications. Real time onsite detection of trace explosives in a
complex environment still remains as a difficult challenge. There is an urgent
demand for further improvements to optimize existing SERS substrates to be more
stable, efficient, selective and sensitive especially for field application. The overall
aims of this research focuses on the development of novel SERS-active substrates for
selective binding and sensitive detection of TNT. The work presented in this thesis
centres around the potential discriminative detection of TNT over two closely related
molecules, 2,4-DNT and picric acid (PA), using two different aminothiols assembly
i.e. cysteamine and 6-aminohexanethiol (AHT) for capturing the target molecule.
Though cysteamine on gold has been previously experimented in other studies [69],
their lack of reproducible Raman signal at trace levels prompted us to refine their
methodology plus expand our studies to a new aminothiol assembly not previously
studied, which is AHT. The prospective of SERS as an effective method is explored
for practical real-time TNT detection in aqueous media down to extremely low
quantities of TNT.
More specifically we seek to:
1) Fabricate functionalized gold nanoparticle assemblies and functionalized
nanostructured substrates for effective SERS activity.
2) Exploit the benefits of self-assembled monolayers (SAMs) on gold substrates
as efficient SERS nanosensors for improved detection of trace explosives.
3) Evaluate the potential of different aminothiols, namely cysteamine and 6-
aminohexanethiol, to form Meisenheimer complex for sensitive detection of
TNT.
4) Investigate the selectivity of both recognition molecules towards the target
analyte, TNT, over possible co-existent interfering molecules.
Introduction 33
5) Compare the efficiency between cysteamine and AHT for functionalization of
SERS substrates and its subsequent detection of TNT by SERS.
6) Compare and contrast between the sensitivity and reproducibility of SERS
measurement from colloidal gold nanoparticles versus nanostructured gold
substrates for the SERS detection of TNT.
7) Assess the capability of the methodology with real life samples for forensic
and environmental applications.
1.11 SAMPLES AND THEIR RAMAN CHARACTERISTICS
The following section describes a few features and chemical structures of the
energetic materials that were utilized throughout this study. The research mainly
focused on detection of 2,4,6-trinitrotoluene (TNT) with interference study
experimented on 2,4-dinitrotoluene (2,4-DNT) and picric acid (PA). All three
samples are typical nitroaromatic explosives. The widespread use and volatility of
nitroaromatics cause them to be an important general class of explosives compounds
for detection. These explosives are not only explosive in nature, but also have several
health and safety concerns. This highlights a growing need to continuously track
explosive compounds to help reduce fatalities and massive damage. The reference
spectra of these substances were acquired by conventional Raman spectroscopy. The
characteristic vibrational modes and their assigned Raman frequencies based on
previous studies are also presented in Figure 1.6-1.8 and Table 1.2-1.4 respectively
[28, 34, 101-104].
2,4,6-trinitrotoluene (2,4,6-TNT)
TNT is a widely used explosive especially in landmines. It is a secondary explosive
that does not occur naturally in the environment. Long-term exposure to TNT may
lead to anaemia, skin irritation, liver function abnormalities and mutagenic effects.
[1, 4, 8, 81]. TNT does not undergo degradation in water and it being a stable
compound means it has long persistence in the environment [105]. Reliable TNT
detection is highly essential preserve human health and homeland security.
34 Introduction
Figure 1.6: Raman spectrum of neat TNT under 785nm excitation wavelength.
Raman frequency (cm-1) Assignment
790-795 C-H out-of-plane bend
820-840 NO2 scissoring mode
1200-1210 C6H2-C vibration
1350-1380 NO2 symmetric stretching
1525-1530 NO2 asymmetric stretching
1610-1620 C=C aromatic stretching
Table 1.2: Characteristic Raman modes of TNT.
2,4-dinitrotoluene (2,4-DNT)
2,4-DNT is a well-known precursor and common degradation product of TNT. It is
also an unavoidable impurity present during manufacturing of military grade TNT
[106]. 2,4 DNT has a similar molecular structure to TNT with only one less NO2
group compared to TNT. Its similar molecular structure means that its Raman
Introduction 35
spectrum has overlapping Raman bands with TNT and thus may cause a false
negative or positive in detection.
Figure 1.7: Raman spectrum of neat DNT under 785nm excitation wavelength.
Raman frequency (cm-1) Assignment
793 C-N stretching
835 NO2 scissoring mode
1207 C-H
1356 NO2 symmetric stretching
1540 C-C stretching
1616 C=C aromatic stretching
Table 1.3: Characteristic Raman modes of DNT.
Picric acid (2,4,6-trinitrophenol)
2,4,6-trinitrophenol, commonly known as picric acid, has a similar structure to TNT,
with an –OH group replacing the –CH3 group in TNT. Dry picric acid is sensitive to
36 Introduction
shock and friction. Picric acid is not only used as a warfare agent, but also commonly
used as industrial chemicals for metal etching and to process dyes. The extensive use
of picric acid has caused significant health and safety concerns as well as
environmental pollutions [107].
Figure 1.8: Raman spectrum of neat picric acid under 785nm excitation wavelength.
Raman frequency (cm-1) Assignment
830 NO2 scissoring mode
1345 NO2 symmetric stretching
1530 NO2 asymmetric stretching
Table 1.4: Characteristic Raman modes of picric acid.
As can be seen from the Raman spectra of all three explosives (Figure 1.6-1.8), the
NO2 symmetric stretching band is the most prominent band. The NO2 scissoring
mode and asymmetric stretching are also common bands between TNT, 2,4-DNT
Introduction 37
and PA. Therefore, if all three molecules were to be present in a sample mixture, it
would be difficult to distinguish, characterize and quantify the exact molecule
present due to the overlapping bands. Thus, positive TNT recognition can be
affected. The similar spectra between similar structured molecules or interfering co-
existent molecules, opens the possibility for false positive detection at trace levels.
This shows the importance of selectivity in detection over molecules with similar
structures. This issue needs to be addressed in order to progress the detection sensor
for integration into real analytical context. The ability to be selective towards the
target molecule particularly for analysis of complex environmental samples, would
meet the needs of widespread, routine detection in the field.
38 Introduction
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 39
Chapter 2: Cysteamine-functionalized
nanoparticles for selective SERS
detection of TNT in forensic
applications
This chapter is made up of the following journal article published in Talanta:
Rapid detection of TNT in aqueous media by selective label free
surface enhanced Raman spectroscopy
Arniza K. M. Jamil, Emad L. Izake, Arumugam Sivanesan, Peter M Fredericks
Nanotechnology and Molecular Sciences Discipline, Faculty of Science and
Engineering, Queensland University of Technology, 2 George St., Brisbane, QLD
4001, Australia. DOI: 10.1016/j.talanta.2014.12.022
40 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 41
42 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 43
2.1 PREFACE
Metal colloids, being the most widely used medium for SERS studies, were utilized
for a label free SERS detection of TNT. Here, we report on the use of cysteamine-
TNT Meisenheimer complexes on citrate-reduced gold nanoparticles for trace
detection of TNT. The easy surface functionalization of gold nanoparticles makes
them the substrate of choice for verification of Meisenheimer complex formation
between cysteamine and TNT. The small size of cysteamine makes it a suitable
candidate for bringing the target molecule closer to the metal surface and thus
enhanced Raman signals. Previous studies by Glembocki et al. [69] on cysteamine
functionalized SERS substrates for detection of TNT showed lack of success. The
major challenge of using cysteamine comes from its preference towards the gauche
configuration, whereby both thiol and amine end of the molecule binds to the metal
surface. This deters the formation of Meisenheimer complexes as the amine moiety is
chemically absorbed to the metal surface. An extension of this methodology was
attempted in this study. The approach used was to form the cysteamine-TNT
complex prior to attachment onto the gold surface. This allowed the amine moiety to
freely recognize and bind to TNT, ensuring formation of Meisenheimer complexes.
A simple and efficient capture mechanism was demonstrated using this approach
with sensitivity down to 1 nM and selectivity over 2,4-DNT and PA.
44 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
2.2 ABSTRACT
We report rapid and ultra-sensitive detection system for 2,4,6-trinitrotoluene (TNT)
using unmodified gold nanoparticles and surface-enhanced Raman spectroscopy
(SERS). First, Meisenheimer complex has been formed in aqueous solution between
TNT and cysteamine in less than 15 minutes of mixing. The complex formation is
confirmed by the development of a pink color and a new UV-Vis absorption band
around 520 nm. Second, the developed Meisenheimer complex is spontaneously self-
assembled onto unmodified gold nanoparticles through a stable Au-S bond between
the cysteamine moiety and the gold surface. The developed monolayer of
cysteamine-TNT is then screened by SERS to detect and quantify TNT. Our
experimental results demonstrate that the SERS-based assay provide an ultra-
sensitive approach for the detection of TNT down to 22.7 ng/L. The unambiguous
fingerprint identification of TNT by SERS represents a key advantage for our
proposed method. The new method provides high selectivity towards TNT over 2, 4-
DNT and picric acid. Therefore it satisfies the practical requirements for the rapid
screening of TNT in real life samples where the interim 24-hour average allowable
concentration of TNT in wastewater is 0.04 mg/L.
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 45
2.3 GRAPHICAL ABSTRACT
46 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
2.4 INTRODUCTION
Nitroaromatic explosives have significant detrimental effects on national security,
ecological environment, and human health [1]. 2,4,6-Trinitrotoluene (TNT) is the
most widely used explosive in industry and military/terrorist activities. TNT is
widely used in landmines and under water blasting, leading to the contamination of
soil and ground water. The treatment of TNT pollution in soil and wastewater is
always difficult and expensive to remedy [108, 109]. Human exposure to TNT
pollution may lead to anaemia and abnormal liver function especially when the
exposure to the pollutant is for prolonged periods of time [110]. To be able to protect
the civilian and military population, it is very important to develop a detection
method which is rapid, simple, sensitive, and can be used by first responders in field
for the identification of TNT threats, whether to the environment or human safety.
A variety of methods are currently available for the detection of TNT [3]. These
methods include fluorescence spectroscopy [111, 112], infrared spectroscopy [113,
114], luminescence spectroscopy [100, 115-119], chromatography [120, 121], ion
mobility spectrometry [122], neutron activation analysis [1], immunochemistry [123-
125], electrochemistry [126-131], molecularly imprinted polymers [99, 132, 133],
immunoassays [134-136]. However, most of these techniques are expensive, non-
portable and are potentially limited by extensive sample preparation [1, 2]. For
example, mass spectrometry requires isolation and purification of the analyte usually
by chromatography.
Surface plasmon spectroscopy has been recently utilized for the detection of TNT in
aqueous solution [137-140]. Also biosensors that utilize the displacement of an anti-
TNT antibody from a surface-confined immune-complex by TNT and the
transduction of the dissociation of the antibody by SPR, have been demonstrated in
the literature [141, 142]. These TNT sensors detect the binding of the molecule by
measuring the change in the refractive index at the interface between a metallic layer
with a fixed receptor and a dielectric medium (e.g. analyte in aqueous solution).
Therefore, they are vulnerable to nonspecific interactions [139, 143].
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 47
Nanomaterial-based electrochemical sensors were also demonstrated for the
detection of TNT. These sensors take advantage of the redox activity of the nitro
groups within the nitro aromatic explosives [144]. Electrode materials such as
graphene [145], polyguanine–SiO2 [146], meso porous SiO2 [147], ordered meso
porous carbon [148] have been used for the detection of TNT in various media.
Recently UV-Vis methods that utilize the use of gold nanoparticles for the detection
of TNT were also demonstrated. In these methods, a primary amine such as 1,2-
ethylenediamine or cysteamine is used to modify the surface of gold nanoparticles
(AuNPs), and the amine capped AuNPs were applied for the detection of TNT by
UV-Vis spectroscopy [15, 64, 66, 67, 149]. In these models the donor-acceptor
interaction between the amine-capped nanoparticles and TNT result in the
aggregation of the nanoparticles and, in effect, a color shift from red to blue. This
visual colorimetric detection is attractive because the detection results can be easily
read out by the naked eye. However these colorimetric methods suffer from the
uncertain identity of the detected analyte since both the aggregation of nanoparticles
and various compounds can influence color change. For example the color change
associated with the detection of TNT is similar to that of codeine, melamine, organic
dithiols and pesticides [149-154]. Also the change in ionic strength or pH of the
sample can induce aggregation and color change. Therefore the above methods carry
the risk of false positive identification especially for samples of complex matrices
[155, 156].
Another nanoparticle-based technique takes advantage of the ability of metal
nanoparticles to quench emission from nearby fluorescent compounds [68]. In this
sensor the presence of the analyte is detected by the change in the intensity of
emission from the fluorescent molecule attached to the detection platform [68, 111,
157-159]. Goldman et al. [160] developed nanoscale sensing assemblies that are
based on fluorescence resonance energy transfer (FRET) for the detection of 0.5 nM
TNT in soil samples. The sensor consisted of anti-TNT specific antibody fragments
attached to a hydrophilic quantum dot (QD) via metal-affinity coordination. A dye-
labelled TNT analogue pre-bound in the antibody-binding site quenches the QD
photoluminescence via proximity-induced FRET. The disadvantage of this model is
that it uses an antibody, which requires carful procedures for functionalization onto
48 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
the QD as well as proper storage of the functionalized QD. Kartha et al. [161]
developed sensitive method for the detection of TNT down to 0.23 ppq using nano
fibres. These nano fibres experience fluorescence quenching in the presence of
nitroaromatic compounds. However, the nano fibre sensor is non-specific to TNT
and responds to other closely related structures such as 2,4,6 trinitrophenol (picric
acid) and dinitrotoluene (DNT) [162]. Recently, Senthamizhan et al. [163] developed
a nano fibre for detecting 1 ppb of TNT in water with high selectivity over 2,4 DNT.
A disadvantage of using optical nano fibres is the fact that surface interactions, such
as van der Waals and Casimir-Polder, may affect both the resonance line shape and
the central position in relation to free-space studies [164]. Ma et al. [165] recently
demonstrated a fluorescent paper sensor that utilizes 8-hydroxyquinoline aluminium
(Alq3)-based fluorescent composite nanospheres for the detection of nitroaromatic
explosives at the ng/mL detection limit.
Surface-enhanced Raman spectroscopy (SERS) has been considered for the sensitive
detection of nitroaromatic explosives [166, 167]. When the analyte molecules are
adsorbed on roughened metal surfaces, their Raman signals are greatly amplified due
to enhanced electromagnetic fields within the immediate vicinity of the metal upon
excitation of plasmon resonances by photon interaction as well as charge transfer
processes between the metal and the adsorbed molecule [1]. The SERS enhancement
effect overcomes the low sensitivity problem inherent in conventional Raman
spectroscopy allowing even single molecule detection [42]. The distinct ability to
obtain molecular recognition of an analyte at very low concentrations in aqueous
environment allows SERS to be a unique technique for ultrasensitive biological and
chemical analysis as well as environmental sensing [168, 169]. Numerous SERS
substrates have been developed for the detection of nitroaromatic explosives [4, 28,
65, 68, 97, 153]. By incorporating a specific chemical moiety on the SERS surface,
one can target the detection of TNT present in a complex sample mixture at ultra-
trace level without the need to physically separate out other interfering species [87].
Molecules such as cysteine, L-cysteine methyl ester, 2-aminothiophenol, p-
aminothiophenol and BSA have been recently used to form Meisenheimer complexes
with TNT and enable the SERS detection of TNT [4, 65, 80, 81, 84, 100].
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 49
In this article, we report the detection of ultra-trace amounts of TNT by surface
enhanced Raman spectroscopy after the rapid formation of a colored Meisenheimer
complex between TNT and cysteamine in aqueous medium. The complex formation
is confirmed by visual inspection and UV-Vis spectroscopy. Using this method, we
were able to directly detect TNT in aqueous media with high selectivity over the
common interfering molecules 2,4-DNT and picric acid. The quantification limit of
the method was found to be 22.7 ng/ L. By combining high sensitivity and selectivity
that are offered in this method, we enhance the accuracy and reliability of our
detection technique.
2.5 EXPERIMENTAL
2.5.1 Reagents
Hydrogen tetrachloroaurate (HAuCl4.4H2O), trisodium citrate, cysteamine
hydrochloride, 2,4-dinitrotoluene (2,4-DNT), 2,4,6-trinitrophenol (picric acid),
sodium hydroxide, acetonitrile, ethanol and methanol were purchased from Sigma
Aldrich (USA). All chemicals and solvents were of analytical grade and were used
without further purification. 2,4,6-trinitrotoluene (TNT) standard (1 mg/ml in 1:1
acetonitrile: methanol) was purchased from Merck (Australia). All dilutions were
made using deionized water (18.2 MΩ.cm) from a Milipore water purification
system.
2.5.2 Gold nanoparticle synthesis
Au NPs were prepared according to the standard citrate reduction method [170, 171].
Typically, 30 mL of 0.1% HAuCl4 was heated to boiling point. Next, 180 µL of 1%
sodium citrate was added to the boiling solution. During heating, the color of the
solution changed from yellow to red. The solution was further heated with stirring
for another 15 minutes before being left undisturbed to cool at room temperature.
The plasmon properties and average size of the prepared gold nanoparticles were
characterized by UV-Vis spectroscopy and high-resolution transmission electron
microscopy (Hr-TEM).
50 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
2.5.3 Preparation of Meisenheimer complex with TNT
Cysteamine stock solution (2 x 10-4 M) was prepared in MiliQ water. In order to
obtain the free base form of cysteamine, the pH of the solution was then adjusted to
pH 8.5 with 0.01 M aqueous sodium hydroxide. To develop the Meisenheimer
complex, equal volumes of TNT and cysteamine standard solutions (2 x 10-4 M)
were mixed together. The mixture is allowed to stand for till its color changes from
colorless to pink then screened by UV-Vis spectroscopy. UV-Vis spectra are
collected at one-minute intervals for 15 minutes in order to determine the optimum
time for the complete formation of the cysteamine-TNT complex.
2.5.4 Detection of TNT by surface enhanced Raman spectroscopy (SERS)
TNT standard solutions in the concentration range 2 x 10-4- 2 x 10-9 M were prepared
by serial dilutions. Equal volumes of a TNT standard solution and cysteamine stock
solution were mixed and allowed to stand for 30 min to develop cysteamine-TNT
complexes at various concretions of TNT. A 9:1 volume ratio of freshly prepared
AuNPs (0.02 nM) and cysteamine-TNT complex were mixed by stirring. The AuNPs
were used as is without any purification. Therefore the TNT concentration range in
the final preparations is between 10-5 –10-10 M. The cysteamine-TNT-AuNPs
mixtures were then deposited onto microscope slides for SERS measurement.
2.5.5 Interference study
To determine the selectivity of the method towards TNT over other nitroaromatic
compounds, standard solutions (2 x 10-4 M) of 2,4-DNT (in ethanol) and picric acid
(in deionized water) were prepared. Equal volumes of 2,4-DNT or picric acid and
cysteamine stock solution are mixed and allowed to stand for 30 minutes. The
mixtures were then screened by UV-Vis spectroscopy. For SERS measurements,
equal volumes of 2, 4-DNT or picric acid (2x 10-5 M) and cysteamine stock solution
were mixed and allowed to stand for 30 minutes. 100 µL of the DNT/cysteamine or
picric acid/cysteamine solution was then mixed with 900 µL AuNP colloid and the
mixture screened by SERS.
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 51
2.5.6 Instrumentation
UV-Vis measurements were carried out in the wavelength range 250 nm-800 nm
using the Agilent Cary 60 UV-Vis spectrometer (Agilent Technologies, USA). The
gold nanoparticles were characterized using the JEM-2100F Transmission Electron
Microscope. Raman spectra were collected on the Renishaw inVia Raman
Microscope using NIR excitation. A 785 nm laser beam was used for excitation. The
applied laser power was 0.1% of maximum power of the laser source (450 mW).
Spectra were collected using a 50x objective lens over a wavelength range from 500
cm-1 to 2000 cm-1 using 10 accumulations of 10 second exposure times each for a
minimum of six independent measurements.
2.6 RESULTS AND DISCUSSION
2.6.1 Development of Meisenheimer complex
Cysteine and cysteamine have high affinity to form Meisenheimer complexes with
TNT. In this work we choose cysteamine as small molecule, of only 2 carbon atoms,
to interact with TNT and form a donor–acceptor molecule. The small chain length of
cysteamine makes it very suitable capturing molecule for the detection of TNT by
SERS since the TNT moiety of the cysteamine-TNT complex will be very close to
the surface of the metal substrate used in the SERS testing. Depending on the pH of
the cysteamine solution, it may predominantly exist in one of three ionic forms: the
positively charged form (HS-CH2-CH2-NH3+), the zwitterionic form ( ¯S-CH2-CH2-
NH3+), and the negatively charged form ( ¯S-CH2-CH2-NH2) [88]. In a neutral
aqueous medium (pH 7), the zwitterionic form of cysteamine interact with the water
molecules so that there are two hydrogen bonds with the cysteamine molecule, that
is, sulphur atom to hydrogen atom of water and from oxygen to hydrogen of the
positively charged amino group. As a result the cysteamine molecule may adopt to a
gauche confirmation where the sulphur and amino groups are engaged in a cyclic
structure via a water molecule bridge [88]. Therefore the cysteamine molecule at pH
7 is incapable of forming a stable Meisenheimer complex with TNT than when it is
in an alkaline medium (pH > 7) where the non-cyclic form ( ¯S-CH2-CH2-NH2)
predominates. The development of 1:1 Meisenheimer complex between alkaline
cysteamine solution (pH 8.5) and TNT is indicated by the color change from
52 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
colorless to pink upon the addition of aqueous cysteamine to the TNT solution
(Figure 2.1). The UV-Vis spectrum of the formed complex is depicted in Figure 2.2a.
As shown in the figure, the formation of the cysteamine-TNT complex results in the
development of a new band at 520 nm. Figure 2.2b depicts the development of the
cysteamine-TNT complex with time. As indicated by the figure, the complex is
formed rapidly, which is evident by the increase in the signal intensity at 520 nm.
The change in the absorbance of the 520 nm band becomes insignificant after 11
minutes which indicates the complete formation of the complex.
Figure 2.1: Development of 1:1 cysteamine-TNT Meisenheimer complex in aqueous
medium at pH 8.5.
The Meisenheimer complex between an electron-deficient nitroaromatic compound
and an electron-rich amine is a resonance-stabilized intermediate structure formed
during a nucleophilic aromatic substitution reaction. For TNT, the three nitro groups
act as strong electron-withdrawing groups that activate the molecule to nucleophilic
substitution [82, 86-88, 170, 171]. Therefore in the presence of cysteamine (electron-
rich nucleophile), a donor-acceptor complex is formed by the addition of a lone pair
of electrons from cysteamine to the aromatic ring of TNT [25, 172]. Figure 2.3
depicts the mechanism for the cysteamine-TNT complex development in aqueous
solution. Stabilization of the developed σ anion complex is through resonance
stabilization of the three electron-withdrawing nitro groups [4]. The nucleophilic
addition at the ring carbon of TNT leads to the formation of a highly conjugated
anion that absorbs light in the visible range and, therefore, develops a pink color as
well as the absorption band at 520 nm [25, 78, 172].
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 53
Figure 2.2: (a) UV-Vis spectrum of the aqueous 1:1 Meisenheimer complex between
cysteamine and TNT and (b) the development of the complex with time.
54 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
Figure 2.3: Mechanism for the formation of 1:1 cysteamine-TNT Meisenheimer
complex.
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 55
2.6.2 SERS detection of TNT
For the SERS fingerprinting and quantification of ultra trace amounts of TNT in
aqueous solutions, we utilized label-free gold nanoparticles (AuNPs). We elected
gold over silver since the later is exposed to oxidation in aqueous media, which
would lead to loss of the SERS effect [173]. The plasmonic properties of the bare
AuNPs were characterised by UV–Visible spectroscopy. The nanoparticles exhibited
a characteristic absorption band at 530 nm. Figure 2.4 depicts the TEM measurement
of the nanoparticles and the size distribution curve. The nanoparticles were found to
be of semi-spherical shape with an average size of 58 ± 9.4 nm.
Figure 2.5 depicts the UV-Vis spectrum of bare AuNPs and AuNPs reacted with
Meisenheimer complex. There was no visible absorbance in the range of 600–800
nm in the absence of cysteamine-TNT complex. A new absorption band around 700
nm arose in the presence of the Meisenheimer complex, which is attributed to the
absorbance of aggregated AuNPs. For the SERS detection of TNT, the analyte was
reacted with cysteamine with TNT, to develop a Meisenheimer complex that is then
immobilized onto the AuNPs colloid (section 2.4.4). The aggregation of AuNPs
causes the nanoparticles to become at a much closer distance from each other and,
therefore, their SPR overlap to form significant electromagnetic field (hot spots).
When an analyte molecule (cysteamine-TNT complex) that lies in the vicinity of the
formed hotspots is excited by incident laser beam, it generates weak Raman
radiation. However, this weak radiation experience very significant enhancement due
to the strong electromagnetic field of the hot spots. This causes the Raman signal to
be amplified by orders of magnitude and a surface enhanced Raman spectrum of the
cysteamine-TNT is acquired [33]. The SERS spectrum of 10-6 M TNT is depicted in
Figure 2.6. As indicated by the figure, the diagnostic spectral lines of TNT at 793
cm-1 (C-H out-of-plane bend), 826 cm-1 (ring deformation), 1215 cm-1 (C-H ring
bend and in-plane rocking) 1350 cm-1 (symmetric nitro stretching) and 1613 cm-1
(2,6-NO2, asymmetric ring stretch) were detected with excellent match to the TNT
reference spectrum [4, 89, 174]. This unambiguous fingerprint identification of TNT
represents a key advantage for our proposed method.
56 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
Figure 2.4: (a) TEM and (b) size distribution of unmodified gold nanoparticles
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 57
Figure 2.5: UV-Vis absorption spectra of AuNP and AuNP-Meisenheimer complex.
Figure 2.6: Raman spectrum of 10-6 M TNT. The TNT solution was first reacted
with cysteamine then with unmodified AuNPs for SERS detection.
58 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
For the SERS quantification, various concentrations of TNT were screened with each
measurement repeated 6 times (n = 6). Figure 2.7a, depicts the Raman signal at
1350 cm-1 of TNT within the concentration range 10-5- 10-10 M (spectra were
normalized and background subtracted for comparison purposes). Figure 2.7b depicts
the relationship between the Raman signal intensity @1350 cm-1 and the
concentration of TNT solution. The relationship between TNT concentration and the
SERS signal intensity (Figure 2.7b) is linear within the concentration range 10-7 M –
10-10 M TNT (R2 =0.9866) and the lower limit of TNT quantification was found to be
0.1 nM (22.7 ng/L).
Figure 2.7: Quantification of ultra trace amounts of TNT in aqueous solution
(a) Raman signal of TNT @ 1350 cm-1,
(b) the relationship between signal intensity and concentration. The error bars
indicate the standard deviation from 6 measurements.
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 59
The success of the proposed method in detecting very low concentrations of TNT is
underlined by our approach to develop the cysteamine-TNT complex in solution
prior to its immobilization onto the AuNPs for SERS measurements. Glembocki et
al. [69] attempted the detection of TNT using cysteamine-coated gold and silver
substrates where they developed a monolayer of the amine onto the metal surface
then used the modified substrate to detect TNT. However their detection strategy
faced serious challenges that made the SERS detection of TNT almost impossible
and non-reproducible. They attributed the unreliable measurements in their
experiment to the fact that cysteamine molecules on gold surface exist mainly in a
gauche configuration. In this configuration the cysteamine binds to the metal surface
from both the thiol and amine ends [175]. This causes the cysteamine molecule to
bend parallel to the metal surface and the amino group of cysteamine is no longer
able to interact with TNT. Therefore the Meisenheimer complex cannot be formed
and the SERS detection of TNT fails. To the contrary, our approach allows the amino
groups to react first with TNT in solution and the resulting complex, when added to
the AuNPs, can only bind to the gold surface via the free thiol group of the
cysteamine moiety to form a stable Au-S bond [176].
Moskovits [177] and Creighton [178] explained that the molecule’s Raman
vibrations that have larger component of polarizability in the direction normal to the
surface experience high SERS enhancement by the substrate. On the other hand, the
least enhanced modes are those whose Raman tensor components involve the two
axes in the plane of the surface [179]. Therefore, the large SERS enhancement of the
TNT Raman bands and the low quantification limit that are observed in our SERS
measurements may be attributed in part to the immobilization of the TNT molecule,
via the cysteamine linker, in an upward position with its molecular plane almost
vertical with respect to the metal surface [175, 179].
A recent report indicated the detection of TNT by gold meso flower shaped
nanoparticles (MFs). The authors indicated the detection limit of TNT per a single
MF to be 0.015 zeptomoles [100]. However, a 100 ng/L TNT solution was
effectively used for the measurement. In addition, under colorimetric conditions the
LOD of the reported method was found to be higher than 44 μM [66]. A more recent
report indicated the detection of 1 pM TNT using dumbbell nano probes [1].
60 Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications
However, the authors used a complex technique for the synthesis of the nano probes.
In addition, the analysis involves multi-step process that requires 24 hours for the
development of a Meisenheimer complex with L-cysteine and an additional 24 hours
for the development of dumbbell hot spots for the SERS detection of TNT.
Therefore, the overall time of 48 hours renders the TNT detection by this method as
impractical. Despite the higher detection limit of our proposed method (22.7 ng/L),
the present method satisfies the practical requirements for the rapid screening of
TNT in real life samples where the interim 24 hour average allowable concentration
of TNT in wastewater is 0.04 mg/L [161, 180, 181].
2.6.3 Interference study
In order to determine the selectivity of the proposed method towards TNT detection
over other nitroaromatic compounds, we investigated the interaction between 2,4-
DNT and picric acid with cysteamine. The structure of picric acid closely resembles
that of TNT. Picric acid is also used for metal etching and in the dye, leather, and
glass industries. Trace amounts picrate anion can exist in water and soil as a pollutant
[107]. 2,4-DNT is found to be an unavoidable impurity in military grade TNT [106].
Therefore it is very important to confirm the selectivity of the analytical method
towards TNT over picric acid and 2,4-DNT in order to avoid false identification.
For the interference study, equal volumes of cysteamine stock solution and 2,4-DNT
and picric acid were reacted under the given experimental conditions (section 2.4.5).
The resulting mixtures were examined by visual inspection, UV-Vis and SERS. The
visual inspection did not indicate any color change after 30 minutes of the reaction.
Also the UV-Vis spectra of the 2,4-DNT/cysteamine and picric acid/cysteamine
mixtures did not show any change to the absorption spectra of 2,4-DNT or picric acid
(Figure 2.8a). The SERS spectra of the cysteamine/2,4-DNT and cysteamine/picric
acid mixtures are depicted in Figure 2.7b and indicate the absence of the
characteristic symmetric nitro stretching Raman signal of nitroaromatic compounds
@ 1350 cm-1. The absence of the band @ 1350 cm-1 is attributed in part to the
formation of a passive cysteamine monolayer onto the gold nanoparticles [175, 182].
This layer neither allows for the adsorptions of 2,4-DNT or picric acid onto the gold
surface nor the formation of a Meisenheimer complex and hence the absence of the
Cysteamine-functionalized nanoparticles for selective SERS detection of TNT in forensic applications 61
SERS signal. These results are also in agreement with the reported literature on the
inability of 2,4-DNT and picric acid to form stable Meisenheimer complexes with
primary amines [4, 15, 64, 66, 158].
Figure 2.8: (a) UV-Vis absorption spectra and (b) SERS spectra of cysteamine/TNT,
cysteamine/DNT and cysteamine/picric acid mixtures. The final TNT concentration
for the UV-Vis measurements was 10-4 M and for the SERS measurements 10-6 M.
The inset depicts the reference UV-Vis spectra of 2,4-DNT and picric acid.
2.7 CONCLUSION
Herein we present a simple and rapid approach to the detection of trace amounts of
TNT in aqueous media. The new method involves the development of a cysteamine-
TNT Meisenheimer complex in solution. The TNT anion in the developed complex
is highly stable due to resonance stabilization through the three NO2 groups of TNT.
Due to the high conjugation in the cysteamine-TNT system, the complex develops a
pink color in solution as well as a characteristic UV-Vis absorption band at 520 nm.
The immobilization of the formed complex onto the AuNPs surface takes place
62
rapidly via the formation of a favoured Au-S bond between the cysteamine moiety of
the complex and the gold surface. Therefore a self-assembled monolayer is formed,
within 30 minutes, onto the gold surface to bring the TNT molecule in an upward
position to the enhancing metal surface. To prevent cysteamine from adopting the
gauche transformation, the Meisenheimer complex between cysteamine and TNT
was carried out first. This initial step safeguards the amine moiety of cysteamine
from binding to the gold surface.
The new method also demonstrated selectivity towards TNT over 2,4-DNT and
picric acid. This was verified experimentally through visual color change of complex
formation, UV-Vis and SERS data. Sensitive SERS detection down to 22.7 ng/L was
demonstrated and it satisfies the practical requirements for the rapid unambiguous
screening of TNT in real life samples where the interim 24 hour average allowable
concentration of TNT in wastewater is 0.04 mg/L. The sensitivity and selectivity of
this new sensor can be expanded for use in rapid screening of TNT residues in soil,
water, on metallic surfaces (e.g. implements used to synthesis TNT and prepare
improvised device, car carrying a bomb), on suspected packages, post explosion
debris (crime scene investigation).
Reproducible nanosensor for SERS detection of TNT in environmental applications 63
Chapter 3: Reproducible nanosensor for
SERS detection of TNT in
environmental applications
This chapter is made up of the following journal article published in Analytical
Methods:
A homogeneous surface-enhanced Raman scattering platform for
ultra-trace detection of trinitrotoluene in the environment
Arniza K. M. Jamil*, Emad L. Izake, Arumugam Sivanesan*, Roland Agoston,
Godwin A. Ayoko
Nanotechnology and Molecular Sciences Discipline, Faculty of Science and
Engineering, Queensland University of Technology, 2 George St., Brisbane, QLD
4001, Australia.
*A.S. and A.K.M.J contributed equally to the manuscript
64 Reproducible nanosensor for SERS detection of TNT in environmental applications
Reproducible nanosensor for SERS detection of TNT in environmental applications 65
66 Reproducible nanosensor for SERS detection of TNT in environmental applications
Reproducible nanosensor for SERS detection of TNT in environmental applications 67
3.1 PREFACE
As a continuation of the study in Chapter 2, this chapter focuses on improving the
metal substrate used by overcoming the issues related to the use of nanoparticles.
The previous study showed the success of simple determination of Meisenheimer
complex formation between cysteamine and TNT to facilitate the sensitive and
selective detection of TNT with gold nanoparticles. The uncontrolled aggregation of
nanoparticles however, limits its application for reproducible SERS signals from
spot-to-spot and substrate-to-substrate. Lack of reproducibility comes from the
random distribution of hot spots. The inconsistent measurements obtained brought
about the aims to improve the methodology by addressing the issue of reproducibility
as well as an added applicability to environmental samples. Nanostructured gold
substrate was utilized in place of nanoparticles for a more efficient substrate for
reproducible, sensitive and selective detection of TNT. The closely packed
nanostructure introduces a more homogenous distribution of hot spots for enhanced
Raman signals. A similar methodology to the previous chapter was used, whereby
the formation of Meisenheimer complex in aqueous solution was performed prior to
placing the sample onto the metal substrate for SERS detection. In this work, we
have shown that by optimizing the SERS substrate, the overall SERS intensity can be
improved. The closely packed gold nanostructures demonstrated better efficiency in
terms of reproducibility and sensitivity in comparison to colloidal gold nanoparticles.
The remarkable sensitivity down to 100 fM can be attributed to the highly ordered
structure of the substrate, resulting in nanogaps for efficient SERS enhancements.
68 Reproducible nanosensor for SERS detection of TNT in environmental applications
3.2 ABSTRACT
A facile and sensitive surface-enhanced Raman scattering substrate was prepared by
controlled potentiostatic deposition of closely packed single layer of gold
nanostructures (AuNS) over flat gold (pAu) platform. The nanometer scale inter-
particle distance between the particles resulted in high population of ‘hot spot’,
which enormously enhanced the scattered Raman photons. A renewed methodology
was followed to precisely quantify the SERS substrate enhancement factor (SSEF)
and it was estimated as (2.2 ± 0.17) × 105. The reproducibility of the SERS signal
acquired by the developed substrate was tested by establishing the relative standard
deviation (RSD) of 150 repeated measurements from various locations on the
substrate surface. A low RSD of 4.37 confirmed the homogeneity of the developed
substrate. The sensitivity of pAu/AuNS was proven by determining 100 fM 2,4,6-
trinitrotoluene (TNT) comfortably. As a proof of concept on the potential of the new
pAu/AuNS substrate in field analysis, TNT in soil and water matrices was selectively
detected after forming a Meisenheimer complex with cysteamine.
Reproducible nanosensor for SERS detection of TNT in environmental applications 69
3.3 GRAPHICAL ABSTRACT
70 Reproducible nanosensor for SERS detection of TNT in environmental applications
3.4 INTRODUCTION
In recent years, surface-enhanced Raman spectroscopy (SERS) has become a very
sensitive analytical technique for identifying traces of analytes ranging from small
chemicals to macromolecules/microorganisms such as bacteria and viruses [183-
188]. Nevertheless, the sensitivity of SERS technique largely relies on the presence
of a sensitive substrate i.e., a strong enhancing surface for the incident and Raman
photons by plasmon resonance [189-191]. SERS enhancement is a joint effect of
electromagnetic and chemical enhancement, the former is a predominant contributor
to the observed enhancement, in particular on metallic surfaces [189, 190].
Ultimately, a surface plasmon rich metallic surface is the prime requirement for large
SERS enhancement. Noble metal nanoparticles represent the best candidate to vastly
amplify the scattered Raman photons since they have a strong plasmon cloud
extending up to several nanometers away from the surface [41, 192]. Numerous
literature have been published in the past decades in relation to metal nanoparticle
based SERS [192, 193]. However, developing SERS substrate that have high
enhancement and good signal reproducibility especially in the case of solution based
metal nanoparticles involve the challenging task of orderly assembling the
nanoparticles on solid platforms to have a consistent SERS effect [194]. This is
because of the challenging task to orderly assemble the nanoparticles on solid
platforms to have a consistent SERS effect. In addition, metallic nanoparticles are
usually capped with stabilizing agent which may significantly interfere with the
SERS spectrum of the target molecule [195]. Thus, researchers are always searching
for a sensitive, stable, reproducible, easy to prepare and user friendly SERS platform
in order to develop cost-effective SERS sensors that are suitable for practical in-field
rapid detection by handheld Raman devices.
In fact, close packing of metal nanoparticles within nanometer space is one of the
vital requirements for an efficient SERS substrate that delivers enormous SERS
enhancement. This is because the nanogaps between particles act as hot spots where
an analyte molecule experiences strong magnetic field and subsequent Raman signal
enhancement [196, 197]. The preparation of a highly sensitive SERS substrate by
electrodeposition is one of the simple, cost effective and efficient approaches that can
overcome many of the challenges and setbacks that are associated with the task of
Reproducible nanosensor for SERS detection of TNT in environmental applications 71
developing closely packed layer of noble metal nanoparticles on solid platform [58,
185]. The advantage of an electrodeposition method is that the reduction of metal
ions, and subsequent deposition of atoms to form particles, can be precisely
controlled by fine-tuning the applied potential and deposition time along with the
right choice of electrolyte [58, 185]. Therefore critical parameters such as particle
size, inter-particle distance and layer thickness can be precisely controlled by
electrodeposition. Further, this method produces bare particles without any capping
agent, which eliminates potential interference from the capping agent Raman
signature. In this work, we applied a simple and quick procedure for
electrodeposition of closely packed single layer of gold nanostructures (AuNS) over
a flat gold surface (pAu) [58]. We preferred gold as the SERS amplifier since it is
more stable than silver in open-air environment. In addition, AuNS are good Raman
photon enhancers under NIR excitation that is used in portable Raman devices.[198]
Further, SERS substrate enhancement factor (SSEF) of the present pAu/AuNS
surface has been precisely quantified by established renewed approach [58, 199].
The explosive, 2,4,6-trinitrotolune (TNT), is commonly used for industrial, military
and recently in global terrorist activities [160]. Therefore, rapid in-field identification
of TNT in soil and water is not only important to forensic and national security
departments but also essential to improve the water and air quality. We recently
demonstrated the synthesis of colored cysteamine-TNT complex in aqueous media
and its use for the detection of TNT by SERS [67, 200]. We also demonstrated that
cysteamine in alkaline aqueous medium is not capable of forming a similar complex
with two important nitroaromatic compounds, 2, 4-dinitrotoluene (2,4-DNT) and
picric acid (PA), which are known to co-exist with TNT in environment. Both of
these compounds are also known to have Raman signatures that are similar to TNT
[200]. Therefore, the ability of detecting TNT without interference from these
compounds confirms the selectivity of the cysteamine-TNT complex for the SERS
detection of TNT.
In our previous work, we used citrate capped gold nanoparticles as the SERS
platform [200]. The disadvantages of using gold nanoparticles were underlined by
the high detection limit (1 nM) of TNT and limited reproducibility of the TNT SERS
signals. These disadvantages are attributed in part to the high polydispersity of the
72 Reproducible nanosensor for SERS detection of TNT in environmental applications
metallic nanoparticles as well as the use of a drop-dry technique. In this technique
the nanoparticles, after interaction with the cysteamine-TNT complex, were dropped
onto a metallic surface and dried. This led to nonhomogeneous distribution of
hotspots and hence the poor reproducibility of the SERS spectra. Therefore, in this
work we aim at using the new pAu/AuNS substrate in order to achieve a proof of
concept on the ultra-trace and reproducible detection of TNT in environment by
SERS. This proof of concept can be easily extended to the in-field SERS detection of
many other analytes by the pAu/AuNS substrate.
3.5 EXPERIMENTAL SECTION
3.5.1 Chemicals and materials
Hydrogen tetrachloroaurate (HAuCl4.4H2O), trisodium citrate, cysteamine
hydrochloride, sodium hydroxide, acetonitrile, ethanol and methanol were purchased
from Sigma Aldrich (USA). All chemicals and solvents were of analytical grade and
were used without further purification. Synthesis of 1,8,15,22-
tetraaminophthalocyanatocobalt(II) (4α-CoIITAPc) was reported elsewhere [201].
2,4,6-trinitrotoluene (TNT) standard (1 mg/ ml in 1:1 acetonitrile: methanol) was
purchased from Merck (Australia). All dilutions were made using deionized water
(18.2 MΩ.cm) from a Millipore water purification system. All other chemicals were
of Analytical grade. The chemicals were used without any further purification.
Polishing slurries and pads (Microcloth®) were purchased from Buehler, Germany.
Polycrystalline gold discs (Au) having a geometric area of 0.502 cm2 and platinum
wire (A & E Metals, Australia) were respectively, used as working and counter
electrode. Dry leakless electrode (DRIREF-2, World Precision Instruments, USA)
was used as a reference electrode.
3.5.2 Instrumentation
All electrochemical experiments were carried out in Autolab PGSTAT204
potentiostat with a custom-made three-electrode cell setup. All Raman measurements
were performed using the Renishaw InVia Raman microscope equipped with 785 nm
laser line as excitation source. Spectra were collected using a 50x and 5x objective
Reproducible nanosensor for SERS detection of TNT in environmental applications 73
lens over a wavelength range from 500 cm-1 to 2000 cm-1 using a laser power of 1
mW for 10 seconds (3 accumulations). For each cysteamine-TNT spectrum, 25
spectra were randomly recorded over the entire surface and averaged. SEM
measurements were performed using Zeiss Sigma VP Field Emission Scanning
Electron Microscope with an accelerating voltage of 5 kV under high vacuum.
3.5.3 SERS substrate preparation
The synthesis of the pAu/AuNS substrate was carried out using our recently reported
methodology with few modifications [58, 199]. Au disc electrodes were manually
mirror-polished with alumina slurries of sequentially decreasing particle sizes (0.5
µm, 0.05 µm and 0.02 µm). After each step of polishing, the electrodes were
immersed in Millipore water and subsequently sonicated in an aqueous ultrasonic
bath for 15 minutes in order to remove the physically adsorbed alumina particles
from the electrode surface.
AuNS were developed over flat Au surface (pAu/AuNS) by potentiostatic deposition.
Prior to the deposition of gold nanostructures, the Au discs were cleaned by
immersing into piranha solution for 10 minutes (3:1, 98% H2SO4 / 30% H2O2) and
subsequent thorough washing with copious amount of Millipore water (warning:
piranha solution is very corrosive and must be handled with extreme caution; it reacts
violently with organic materials and should not be stored in tightly closed vessels).
After cleaning the Au disc, a three-electrode electrochemical cell was filled with the
solution of 4 mM HAuCl4 in 0.1 M HClO4 and subsequently purged with high pure
argon gas for 30 minutes to remove oxygen from solution. A potential of -80 mV
was applied for 400 seconds and then the electrode was removed from the solution
and subsequently washed with Millipore water to remove other ions from the surface.
The electrode was then dried under a stream of nitrogen gas and used as a SERS
substrate.
3.5.4 Preparation of cysteamine- TNT complex
Cysteamine-TNT complex was prepared according to the recently developed
procedure [200]. Briefly, cysteamine stock solution (2 x 10-4 M) was prepared in
74 Reproducible nanosensor for SERS detection of TNT in environmental applications
MilliQ water and the pH adjusted to 8.5. Equal volumes of TNT and cysteamine
having the desired concentration were mixed together. The mixture was then allowed
to stand for 15 minutes until the pink color of the Meisenheimer complex is fully
developed.
3.5.5 Detection of TNT by SERS
After the preparation of the cysteamine-TNT complex at the desired concentration,
100 µL aliquot of the complex was then dropped over the pAu/AuNS surface and
allowed to stand for 30 minutes in order to form a self-assembled monolayer of the
cysteamine-TNT complex through the free thiol group of cysteamine. Subsequently,
the substrate was thoroughly washed with copious amount of water to remove the
unbound and physically adsorbed cysteamine-TNT complex from the surface. The
substrate was then dried in a stream of nitrogen gas and subsequently used for SERS
measurement to identify the TNT. A laser source @785 nm was used in this work
due to its availability. This NIR excitation wavelength has been utilized in other
SERS studied [67, 188, 200].
3.5.6 Identification of TNT in soil
A known concentration of TNT was spiked into soil and subsequently extracted into
water medium by washing the TNT treated soil with water and centrifugation. The
extracted aqueous TNT was then interacted with cysteamine at pH 8.5 to form the
Meisenheimer complex. Finally, the TNT-cysteamine complex was dropped onto
pAu/AuNS substrate and allowed to stand for 30 min. The substrate was then washed
with millipore water, dried and used for SERS measurements.
3.6 RESULTS AND DISCUSSION
3.6.1 pAu/AuNS substrate preparation and characterization
The preparation of various nanostructures of gold via electrodeposition and its
subsequent SERS application has been demonstrated in literatures [202, 203].
However, the majority of the reported substrates were non-homogenous and their
Reproducible nanosensor for SERS detection of TNT in environmental applications 75
surfaces lack highly ordered geometry. Therefore, in order to overcome these
problems and to develop a homogenous surface, in the present work closely packed
gold nanostructures were formed over the Au surface. We optimized the
electrodeposition conditions such as gold chloride concentration, applied potential,
deposition time and electrolyte. Figures S1 (A-C) shows the SEM of nanostructured
gold substrates that are prepared at different applied potentials (-20 mV, -150 mV,
and -300 mV). The figure clearly indicates the non-homogenous surfaces of the
substrates. Figure 3.1 depicts the SEM of the nanostructured substrate, prepared at
-400 mV, under different magnifications. It is clear from the figure that the substrate
prepared at -400 mV has a more homogenous surface compared to the other
substrates prepared at- 20 mV, -150 mV, and -300 mV (S1 A-C). Thus, the
fabrication conditions were optimized to 4 mM HAuCl4 in 0.1M HClO4 with the
applied potential of -400 mV for 400 seconds.
The close packing of single layer gold nanostructures within nanometer scale inter-
particle distance is very critical for enormous SERS enhancement via hot spots and
also for the reproducibility of the SERS signal over the entire surface [196, 197]. The
SEM image of pAu/AuNS surface under wider magnification (Figure 3.1A) reveals
that the deposition of AuNS is almost uniform over the entire surface. Even under
100 micron magnification, the particle coverage was uniform and almost defect-free.
We obtained consistent SEM pictures over the 8 mm diameter surface, which clearly
demonstrates that the electrodeposition method produced AuNS over the entire
surface. The size of AuNS was in the range of 10-100 nm (Figure 3.1C and D).
Although the sizes of the AuNS are not highly monodisperse, they are ideal for
delivering homogeneous SERS signal when the spot diameter of the focused laser is
micron scale as indicated by Figure 3.2. A closer look at the magnified SEM image
(Figure 3.1C and D) reveals that the AuNS are deposited almost as a single layer.
Furthermore, the AuNS are closely packed having nanometer scale inter-particle
space which led to creating vast number of hot spots. Therefore, despite the
polydispersed particle size (10-100 nm), we believe that the homogeneity in particle
coverage and inter-particle distance significantly reduces the variations of the SERS
signal at different locations for a micron-scale diameter focused laser spot (vide
infra).
76 Reproducible nanosensor for SERS detection of TNT in environmental applications
At this juncture, it is worth focusing briefly on our choice of polished Au as the
underlying platform. For the deposition of AuNS, choosing the same material (i.e.,
Au) as the underlying support allows several particle initiation spot, which
subsequently leads to smaller AuNS size and high packing density. In addition to the
hot spot effect, a possible coupling between the propagating surface plasmon
polariton (SPP) of the underlying polished Au surface and surface plasmon
resonance (SPR) of the AuNS (i.e., SPP–SPR coupling) [58, 204] would lead to high
SERS enhancement in comparison to the conventional SERS enhancement from
nanoparticles only. In order to quantify the SERS substrate enhancement factor
(SSEF), we applied a renewed electrochemical approach (see supplementary
information). The SSEF of pAu/AuNS surface has been precisely quantified and was
found to be (2.2 ± 0.17) × 105 [58, 199].
Figure 3.1: (A-D) SEM images of pAu/AuNS surface, prepared at -400mV, under
different magnifications.
A B
C D
Reproducible nanosensor for SERS detection of TNT in environmental applications 77
The homogeneity of the substrate to deliver consistent SERS signal was
demonstrated by randomly recording 150 surface-enhanced resonance Raman
(SERRS) spectra of 4α-CoIITAPc over the entire pAu/AuNS surface. The relative
standard deviation (RSD) of the 756 cm-1 peak (Figure 3.2A) was found to be 8.23%
when a 50x objective is used to focus the laser beam (spot diameter = 0.64 µm;
illuminated area = 0.32 µ2, see supplementary information). However, the RSD
decreases down to 4.37% when a 5x objective is used where the diameter of the
focused laser beam is around 3.99 µm with a focusing area of 12.56 µ2 (Figure 3.2B).
The observed decrease in RSD with increase in laser focusing area (5x objective has
39.25 times higher illumination area than 50x objective) is justified as the SERS
signal becomes highly averaged with the increased focus of the objective. In our
experiment, when using 5x objective, the SERS substrate is actually screened under
wide area illumination (WAI) setting in comparison to 50x objective. This WAI
setting allows for increased focusing area and, in effect, contributes to the enhanced
reproducibility of the SERS signal [205, 206]. Thus, the obtained RSD of 4.37% for
the laser beam illumination area of 12.56 µ2 clearly confirms that the surface has
high efficiency for reproducible SERS performance, especially under wide area
illumination settings. The above results clearly indicate that the homogeneity in
particle coverage, as a single layer, and the uniformity in particle packing density
leads to with consistent hot spot distribution and homogeneous SERS signal.
78 Reproducible nanosensor for SERS detection of TNT in environmental applications
Figure 3.2: A series of SERRS spectra of 4α-CoIITAPc randomly collected over the
entire 8 mm diameter pAu/AuNS disc using (A) 50x objective with a laser focusing
area of 3.14 µ2 and (B) 5x objective with a laser focusing area of 132.67 µ2.
Reproducible nanosensor for SERS detection of TNT in environmental applications 79
3.6.2 SERS detection of TNT
Self-assembled monolayer of cysteamine-TNT Meisenheimer complex was formed
over pAu/AuNS surface according to the experimental procedure. The advantage of
the cysteamine-TNT Meisenheimer complex is that the cysteamine moiety of the
complex is acting as a binding molecule that forms a strong Au-S bond with the
nanostructured surface of the substrate [199]. Due to the very small length of the
cysteamine molecule, the cysteamine-TNT complex is brought to a very short
distance from the SERS substrate. Therefore, the Raman spectrum of TNT
experiences a high enhancement by the SERS substrate. In addition cysteamine was
proven by our group to selectively bind with TNT to form a Meisenheimer while
other nitroaromatic compounds such as 2,4-DNT and picric acid did not form a
similar complex at pH 8.5 [200]. Figure 3.3A (pink spectrum) depicts the SERS
spectrum of the monolayer formed from 1 nM cysteamine-TNT complex. The SERS
spectrum of cysteamine-TNT complex is dominated by C=C aromatic stretching
vibration at 1643 cm-1 due to the vertical orientation of TNT over the pAu/AuNS
surface and the large Raman cross-section of the aromatic ring [4, 103, 207, 208].
The spectrum also depicted a sharp and intense band at 1348 cm-1 corresponding to
NO2 group symmetric stretching vibration [4, 103, 207, 208]. The other significant
bands appearing at 1292 cm-1, 1253 cm-1 and 797 cm-1 correspond to aromatic ring
stretching, C6H2-C vibration and C-H out of plane bending, respectively [4, 103, 207,
208]. The SERS spectrum of cysteamine monolayer alone does not show any
significant Raman bands at a concentration of 10 pM. However, it showed significant
Raman bands for monolayer formed at higher concentration (10-5 M), which is
depicted in Figure S2.
In order to demonstrate the suitability of the present Au/AuNS towards SERS
quantification, various concentrations of cysteamine-TNT complex were prepared
and the monolayer modified Au/AuNS substrates were screened under the Raman
microscope. Figure 3.3A shows the SERS spectrum of the cysteamine-TNT
monolayer within a concentration range of 1 nM-100 fM (the spectra were
normalized and background subtracted from the spectra of cysteamine on
pAu/AuNS.). For each concentration, 25 spectra were randomly recorded over the
entire surface and averaged. The band centered at 1348 cm-1 corresponding to NO2
80 Reproducible nanosensor for SERS detection of TNT in environmental applications
symmetric stretching of TNT was used a reference band for TNT quantification. The
SERS signals were found to be monotonically decreasing with the decreased
concentration (Figure S3) while the spectral features remain identical for each
concentration. Even at the concentration of 100 fM, the spectral features of TNT can
still be observed (Figure 3.3A inset). The linear relation between SERS signal
intensity at 1348 cm-1 and the corresponding log concentration of cysteamine-TNT
complex is depicted in Figure 3.3B. The linear relationship between log
(concentration) and SERS intensity was previously demonstrated in the literature [1,
97]. As indicated by the Figure 3.3B, good correlation (R2 = 0.992) was found over a
wide concentration range of TNT (10-9- 10-13 M)
The SERS spectrum of the cysteamine-TNT complex showed good Raman counts
(around 1000 counts as an average) even for a very low concentration of 100 fM
(Figure 3.3A inset). This clearly demonstrates that pAu/AuNS surface is highly
sensitive which is credited to the close packing of AuNS with nanometer scale inter-
particle distance generating several hotspots. The shorter chain length (2 carbons) of
cysteamine may also be responsible for the higher SERS signal of TNT since the
SERS signal decreases by a factor of 10-12 while increasing the distance between the
analyte and Raman enhancing surface [209]. At the low concentration of 100 fM, the
observed SERS spectra would be reported from single hot spots. Thus, SERS
temporal changes using 50x objective may show blinking and, hence, the ultra-high
sensitivity of the SERS substrate [210].
Reproducible nanosensor for SERS detection of TNT in environmental applications 81
Figure 3.3: (A) SERS spectra of decreasing TNT concentration: 1 nM (pink), 100
pM (blue), 10 pM (green), 1 pM (red), 100 fM (black). Inset shows the closer view
of 100 fM spectrum. For each concentration, 25 spectra were randomly recorded
over the entire surface and averaged. (B) Plot demonstrating the linear relation
between the concentration of TNT and Raman intensity @1348 cm-1.
82 Reproducible nanosensor for SERS detection of TNT in environmental applications
3.6.3 Identification of TNT in soil
Identifying TNT in soil sample has demonstrated the practical application of the
nanostructured substrate (see experimental section). Prior to spiking the soil matrix
with TNT solution, the sample has been tested for any pre-existing TNT within the
matrix and the SERS spectrum of the extract confirms the absence of any such pre-
existing TNT. Conversely, after spiking TNT into the soil matrix the subsequently
extracted cysteamine-TNT complex (Figure 3.4; red spectrum) showed almost
identical Raman fingerprints to those of the fresh cysteamine-TNT complex. This
clearly demonstrates that the present methodology can be employed to identify TNT
in field for environmental and forensic applications.
Figure 3.4: Monolayer SERS spectra of bare pAu/AuNS (blue), 10 pM cysteamine
(green), 1 nM cysteamine-TNT (black) and 1 nM soil extracted TNT complexed with
cysteamine (red) on pAu/AuNS surfaces.
Reproducible nanosensor for SERS detection of TNT in environmental applications 83
3.7 CONCLUSION
In conclusion, the present study has simultaneously demonstrated the strength of
electrodeposition method and SERS technique by producing an ultra-sensitive SERS
substrate and identifying 100 fM of TNT, respectively. By identifying 1 nM of TNT
in soil we have demonstrated a proof of concept for the rapid in-field identification
of ultra-trace amounts of TNT by using a small molecule to selectively bind to the
analyte and immobilize it onto our ultrasensitive SERS substrate. Presently, we are
involved in developing disposable SERS strips based on the present electrochemical
approach to be coupled with portable Raman device as a readout approach for in-
field analysis.
3.8 ACKNOWLEDGEMENT
The authors sincerely thank Dr. Llew Rintoul from QUT for his valuable discussion
and support.
84 Reproducible nanosensor for SERS detection of TNT in environmental applications
3.9 SUPPORTING INFORMATION
Calculation of laser spot size
The simplified equation to calculate the diameter of the focused laser spot is
S = 0.61λ / NA
The above equation is extracted from Renishaw notes of the Spectroscopy Products
Division; SPD/PN/088 Issue 1.1 June 2003)
here S = diameter of laser spot size (beam diameter)
λ = wavelength of the laser beam which is 785 nm in the present study
NA = Numerical aperture
50 x objective
NA = 0.75 (Renishaw standard)
Spot size, S = 0.61x 0.785 /0.75 = 0.64 µm
Radius = 0.32 µm
Illuminated area = 0.32x0.32 x3.14 = 0.32 µm2
5 x objective
NA = 0.12 (Renishaw standard)
Spot size S= 0.61 x 0.785/ 0.12 = 3.99 µm
Radius = 2 µm
Illuminated area = 2 x2x 3.14=12.56 µm2
Therefore the illuminated area by the 5 x objective is 39.25 times larger than the 50 x
objective.
Reproducible nanosensor for SERS detection of TNT in environmental applications 85
.
Fig
ure
S1 S
EM
im
ages
of
pA
u/A
uN
S s
ubst
rate
s pre
par
ed a
t (A
) -2
0 m
V,
(B)
-150 m
V a
nd (
C)
-30
0 m
V r
espec
tivel
y.
86 Reproducible nanosensor for SERS detection of TNT in environmental applications
Figure S2 Monolayer SERS spectrum of 10-5 M cysteamine on pAu/AuNS surfaces.
Reproducible nanosensor for SERS detection of TNT in environmental applications 87
Figure S3 Plot demonstrating relation between the concentration of TNT and Raman
intensity @1348 cm-1.
88 Reproducible nanosensor for SERS detection of TNT in environmental applications
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 89
Chapter 4: Aminohexanethiol mixed
monolayer nanosensor for SERS
detection of TNT in the
environment
This chapter is made up of the following journal article published in Sensors and
Actuators B: Chemical:
Molecular recognition of 2,4,6-trinitrotoluene by 6-
aminohexanethiol and surface-enhanced Raman scattering sensor
Arniza K. M. Jamil*, Arumugam Sivanesan*, Emad L. Izake, Godwin A. Ayoko,
Peter M Fredericks
Nanotechnology and Molecular Sciences Discipline, Faculty of Science and
Engineering, Queensland University of Technology, 2 George St., Brisbane, QLD
4001, Australia.
*A.S. and A.K.M.J contributed equally to the manuscript
DOI: 10.1016/j.snb.2015.06.046
90 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 91
92 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 93
4.1 PREFACE
A limitation of the previous work in Chapter 3 is the tendency of the cysteamine
molecule to adopt the gauche transformation, whereby both ends of the molecule
attach to the metal surface. This decreases the amount of molecules that are able to
interact with cysteamine as the recognition molecule. Therefore, there are less or no
free amine groups for the TNT molecule to interact with for formation of
Meisenheimer complex. As a replacement for cysteamine, a different recognition
molecule was studied to increase the efficacy of this technique. The aminothiol
chosen is based on its ability to absorb onto the metal surface in an upright position
leaving the amine moiety free for TNT to attach to.
In this study 6-aminohexanethiol (AHT) was selected, as its longer chain favours the
formation of a packed monolayer on the substrate. This overcomes the issue of the
gauche transformation of cysteamine and thus removes the need for the reaction step
of Meisenheimer complex formation in solution prior to introducing the sample onto
the substrate. In addition, the use of AHT allows the formation of mixed monolayers
with other thiols, in this case, butanethiol (BT), to fill the gaps between adjacent
AHT molecules. Not only does this prevent interfering molecules, it also allows
sufficient space for the target molecule, TNT, to position itself comfortably on the
amine moiety. The feasibility of AHT functionalized nanostructures was attempted in
both soil and water samples.
94 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
4.2 ABSTRACT
2,4,6-trinitrotoluene (TNT) is one of the most commonly used nitro aromatic
explosives in landmine, military and mining industry. This article demonstrates rapid
and selective identification of TNT by surface-enhanced Raman spectroscopy
(SERS) using 6-aminohexanethiol (AHT) as a new recognition molecule. First,
Meisenheimer complex formation between AHT and TNT is confirmed by the
development of pink color and appearance of new band around 500 nm in UV-
Visible spectrum. Solution Raman spectroscopy study also supported the AHT:TNT
complex formation by demonstrating changes in the vibrational stretching of AHT
molecule between 2800-3000 cm-1. For surface enhanced Raman spectroscopy
analysis, a self-assembled monolayer (SAM) of AHT is formed over the gold
nanostructure (AuNS) SERS substrate in order to selectively capture TNT onto the
surface. Electrochemical desorption and X-ray photoelectron studies are performed
over AHT SAM modified surface to examine the presence of free amine groups with
appropriate orientation for complex formation. Further, AHT and butanethiol (BT)
mixed monolayer system is explored to improve the AHT:TNT complex formation
efficiency. Using a 9:1 AHT:BT mixed monolayer, a very low detection limit (LOD)
of 100 fM TNT was realized. The new method delivers high selectivity towards TNT
over 2,4 DNT and PA. Finally, real sample analysis is demonstrated by the extraction
and SERS detection of 302 pM of TNT from spiked soil.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 95
4.3 GRAPHICAL ABSTRACT
96 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
4.4 INTRODUCTION
Trinitrotoluene (TNT), a highly explosive nitroaromatic compound plays a very
significant role in the mining industry and underwater blasting [98]. Moreover, TNT
is also widely used in military and terrorist activities in the form of landmines and
bombs for several decades [211, 212]. Due to this widespread application, TNT
pollution in environment requires ongoing monitoring [213, 214]. In particular, soil
and ground water is sometimes highly contaminated beyond the human tolerant level
[213, 214]. Prolonged exposures of TNT to humans result in anaemia and liver
abnormalities [110, 215]. Further, TNT also act as a carcinogen and leads to cancer
development in living system [216]. Since TNT can easily absorb through the skin
[217], even external contact of humans and other organisms with TNT-contaminated
water and soil may create significant damage to their biological systems.
Over the years, numerous analytical methods have been developed for the detection
of TNT in soil and water [9, 218-221]. However, the in-field and real-time
determination of TNT at the ultra-trace level still face limitations in terms of
sensitivity, selectivity, reproducibility and cost-effectiveness. Surface-enhanced
Raman spectroscopy (SERS) is a powerful analytical technique known for its ultra-
sensitivity and rapidity in sample analysis [185, 209, 222-224]. The miniaturization
of Raman spectrometer into a handheld device has greatly revolutionized the in-field
application of Raman spectroscopy [225]. Richard Van Duyne [226] recently
demonstrated the application of handheld Raman spectrometer for the in-field
forensic analysis of dyed hair by SERS. This study clearly indicates that SERS will
be one of the next-generation analytical tools for the rapid in-field analysis of target
molecules. To realize the rapid in-field detection by SERS, cost-effective, sensitive
and selective substrates are essential [227]. In order to detect TNT in environmental
samples, a cost effective disposable SERS strip that is functionalized with a suitable
binding molecule to selectively recognize TNT over other interfering analytes is
highly desirable. This study focuses on finding a new recognition molecule for TNT
such that the recognition molecule can be attached to the disposable SERS strips
surface. This recognition molecule should selectively bind TNT onto the strip surface
in a suitable orientation to enable a high SERS signal enhancement of the analyte
Raman fingerprint. Further, the recognition molecule should have a higher
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 97
mechanical and thermal stability, which will enhance the shelf life of the sensor
[228]. The functionalized SERS strips will lead to rapid and selective in-field
analysis of TNT trace amounts using handheld Raman device [229].
Several literatures have been reported over the years for sensitive and rapid SERS
based detection of TNT using various nanoparticles and SERS platforms. A wide
range of primary amines has been used for TNT recognition since primary amines
effectively forms Meisenheimer complex with this nitro aromatic compound. For
example, cysteine [4, 81], aminothiophenol [84, 98], BSA [100] and cysteamine
[200, 230] have been used as selective TNT recognition molecules. In the present
work, we have chosen 6-aminohexanethiol (AHT) as a potential selective TNT
recognition molecule. AHT has several advantages over other primary amines that
have been utilized for TNT recognition. These advantages are:
i) The longer chain length of AHT compared to cysteamine provides an
improved long-term stable self-assembled monolayer (SAM) on SERS
substrate. This is due to the enhanced van der Waals lateral interactions
between adjacent AHT chains [228, 231]. This long-term stability of AHT
monolayer on SERS active surface will increase the shelf life of the sensor
[228].
ii) AHT forms a vertically orientated monolayer with amine groups protruding
away from the surface [231-234] and avoids the ambiguity of chelate binding
of both thiol and amine groups onto the gold surface as in the case of
cysteamine [92, 175].
iii) AHT delivers the flexibility in forming a mixed SAM by creating a passive
backfill layer using shorter thiols such as butanethiol (BT). This shorter
backfill molecule also acts as a spacer molecule such that it will create
enough space between the AHT molecules and provides easy access to TNT
to bind, via the AHT bridge, in an almost perpendicular orientation to the
SERS substrate [175, 179, 200, 235]. It has been demonstrated in the
literature that, high SERS enhancement is realized when the molecule’s
Raman vibrations have large component of polarizability in the direction
normal to the SERS surface [177, 178, 200]. On the other hand, the least
enhanced modes are those whose Raman tensor components involve the two
axes in the plane of the surface [179].
98 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
iv) AHT will not form any chelating complex with metal ions, as in the case of
cysteine. This is due to the existence of only one metal ion coordinating
amine group in the AHT molecule. In contrast, cysteine, with both amine and
carboxylic group, has been known for its ability to form chelating complex
with metal ions [96, 236]. This may hinder the amine group of the cysteine
molecule to access TNT, especially in environmental samples that may
contain several metal ions.
v) AHT is a weak Raman scatterer and will show less interference with TNT
signal when compared to other aromatic primary amines [84, 98] used for
TNT recognition such as p-aminothiophenol.
The present manuscript involves the development of a new Meisenheimer complex
between AHT and TNT and its subsequent characterization by UV-Visible and
Raman spectroscopy. SAM of AHT alone and mixed monolayer of AHT:BT are
developed on gold nanostructured (AuNS) SERS substrate. The developed
monolayers are examined for their performance as recognition layers that bind TNT
to the SERS substrate via the formation of selective Meisenheimer complex with the
analyte. Various ratios of AHT:BT are attempted for the mixed monolayer formation
in order to optimize the AHT:BT monolayer for a strong SERS signal that allows for
the detection of ultra-traces of TNT. To conclude, SERS substrate that is modified
with 9:1 AHT:BT mixed monolayer has been employed for the selective and
sensitive detection TNT in soil through the established calibration curve.
4.5 EXPERIMENTAL SECTION
4.5.1 Chemicals and materials
Hydrogen tetrachloroaurate (HauCl4.4H2O), 6-aminohexanethiol hydrochloride
(AHT), butanethiol (BT), cysteamine hydrochloride (CA), 2,4-dinitrotoluene (DNT),
2,4,6-trinitrophenol (picric acid) and sodium hydroxide (NaOH) were purchased
from Sigma Aldrich. 2,4,6-trinitrotoluene (TNT) standard (1 mg/mL in 1:1
acetonitrile:methanol) was purchased from Merck (Australia). All other chemicals
and solvents were of analytical grade and were used without further purification. All
dilutions were made using deionized water (18.2 MΩ.cm) from a Millipore water
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 99
purification system. Polishing slurries and pads (Microcloth®) were purchased from
Buehler, Germany. Polycrystalline gold discs (Au) having a geometric area of 0.502
cm2 and platinum wire (A & E Metals, Australia) were respectively, used as working
and counter electrode. Dry leakless electrode (DRIREF-2, World Precision
Instruments, USA) was used as a reference electrode.
4.5.2 Instrumentation
All electrochemical experiments were carried out in Autolab PGSTAT204
potentiostat with a custom-made three-electrode cell setup. All Raman measurements
were performed using the Renishaw InVia Raman microscope equipped with 785 nm
laser line as excitation source. Spectra were collected using a 50× and 5x objective
lens over a wavelength range from 500 cm-1 to 2000 cm-1 using a laser power of
1 mW for 10 seconds. For each AHT-TNT spectrum, 10 spectra were randomly
recorded over the entire surface and averaged. X-ray photoelectron (XPS)
measurement was performed using Omicron Multiscan scanning probe microscopy.
UV–Vis measurements were carried out in the wavelength range 250–800 nm using
the Agilent Cary 60 UV–Vis spectrometer (Agilent Technologies, USA).
4.5.3 SERS substrate preparation and mixed monolayer formation
The SERS active substrate was prepared by electrodepositon of gold nanostructure
over flat gold surface (pAu/AuNS) in accordance with our previous reported method
[58]. Self-assembled monolayer (SAM) of cysteamine, AHT and mixed monolayer
of AHT and BT was prepared by immersing the substrate in already prepared
appropriate ratio of 1 mM aqueous solution of respective thiols. After 3 h the
substrates were washed with copious amount of water to remove the freely adsorbed
molecules from the electrode surface.
The electrochemical desorption measurements were carried out in 0.1 M KOH at a
scan rate of 0.1 Vs-1. Prior to applying desorption potential, the electrochemical cell
was purged with argon gas for about 10 minutes to completely remove the oxygen
present in the electrolyte solution.
100 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
4.5.4 Preparation of Meisenheimer complex with TNT both in solution and on
surface
AHT:TNT complex was prepared similar to the reported procedure for other primary
amines and TNT [86, 200]. Briefly, AHT stock solution (2x 10-4 M) was prepared in
MilliQ water and the pH is adjusted to 8.5. Equal volumes of TNT and AHT having
the desired concentration were mixed together. The mixture was then allowed to
stand for 15 minutes until the pink color of the Meisenheimer complex is fully
developed. To prepare the Meisenheimer complex directly on pAu/AuNS surface,
few drops of TNT at the desired concentration and optimum pH (8.5) were placed for
about 10 minutes on the SERS substrate modified with AHT monolayer. The
modified pAu/AuNS substrate was then washed with copious amount of NaOH
solution (pH 8.5) to remove any un-complexed TNT from the surface. NaOH was
used to remove excess unbound TNT onto aminohexanethiol modified AuNS to
maintain the pH8.5 for a conducive environment for Meisenheimer complex
formation. The substrate was then dried under a gentle flow of nitrogen and used for
SERS measurements.
4.5.5 Extraction of TNT from soil
A known concentration of TNT was spiked into soil and subsequently extracted into
a water suspension by dispersing the TNT treated soil with water and subjected to
centrifugation. The extracted aqueous TNT was then interacted with the AHT-
modified pAu/AuNS susbtrate to form the Meisenheimer complex. The AHT-TNT
complex was allowed to stand for 10 minutes and then washed with copious amount
of NaOH solution (pH 8.5). The substrate was then dried under a stream of nitrogen
and sent for SERS measurements.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 101
4.6 RESULTS AND DISCUSSION
4.6.1 Characterization of AHT-TNT Meisenheimer complex in solution phase
Meisenheimer complexes are σ-complexes formed by the nucleophilic addition of
electron-rich species, usually amines, to the aromatic system of electron-deficient
aromatic substrates [86, 237]. Due to three strong electron withdrawing nitro groups
in its structure, TNT exists as an electron deficient aromatic system, which readily
forms Meisenheimer complex with amines [4]. AHT is an electron-rich molecule
with its amine group capable of contributing a lone pair of electrons to form a stable
complex with TNT [86, 237]. We carried out the AHT:TNT complex formation at
alkaline medium (pH 8.5). Figure 4.1A shows the UV-Visible spectrum of AHT,
TNT and AHT:TNT mixture at pH 7 and pH 8.5. As expected, the UV-Visible
spectrum (pink spectrum) of AHT:TNT mixture at pH 8.5 showed a new band at 500
nm confirming the formation of Meisenheimer complex. This was adequately
supported by the photograph we took at a higher concentration (Figure 4.1A inset)
where the pink color appeared upon the formation of Meisenheimer complex
between AHT:TNT. On the other hand, AHT:TNT at pH 7 (red spectrum) does not
show any significant intense band at 500 nm which confirms that the alkaline
medium (pH 8.5) is vital for the complex formation between AHT and TNT.
Similarly, TNT alone at pH 8.5 (blue spectrum) does not show any bands at 500 nm.
The development of AHT:TNT complex is vital for the SERS based detection of
TNT using AHT as recognition molecule. In order to identify the saturation time of
the complex formation, the increase of the UV-Vis signal intensity with time was
monitored and recorded (Figure 4.1B). According to the Figure 4.1B, the band
indicative of AHT:TNT complex formation at 500 nm saturates after 10 minutes
where no further appreciable increase in absorbance can be observed.
102 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Figure 4.1: UV–Vis absorption spectra of aqueous 1:1 Meisenheimer complex
between AHT and TNT (A) at neutral and alkaline pH and (B) Absorbance vs time
plot for the formation of Meisenheimer complex between AHT and TNT at pH 8.5.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 103
The formation of Meisenheimer complex between AHT:TNT was further
characterized by solution Raman spectroscopy. Figure 4.2A and B (black) shows the
Raman spectrum of clean AHT in aqueous medium at pH 8.5. AHT showed good
Raman spectrum with characteristic bands in the fingerprint region between 1000
cm-1 and 1650 cm-1. The addition of TNT to AHT at pH 8.5 (red spectrum in Figure
4.2A) showed intense bands at 1349 cm-1 and 1618 cm-1 corresponding to symmetric
NO2 stretching and aromatic ring stretching modes respectively, which was not
readily noticed in the AHT spectra in Figure 4.2A (black spectrum). To confirm the
complex formation between AHT and TNT, we focused our attention to the two
bands at 2800 and 3000 cm-1. According to the literature the bands at 2858 cm-1 and
2910 cm-1 correspond to the symmetric and asymmetric CH2 stretching, respectively
[4, 88, 176]. Among these, the higher frequency CH2 stretching at 2910 cm-1 is very
sensitive to the protonation of the amine moiety and can be used as a vibrational
marker to study the ionization of amine groups [88]. Significant changes in the
electron density of the amine group leads to a change in the intensity of the 2910
cm-1 band [88]. A careful examination of the CH2 stretching region (2800 cm-1-
3000 cm-1) in the SERS spectrum of the AHT:TNT complex (red spectrum in Figure
4.2 B) indicates a change in the band shape in comparison to that of the clean AHT
(black spectrum, in Figure 4.2 B). The 2910 cm-1 band in the AHT:TNT spectrum
(red spectrum) almost disappeared and only a broad band at 2858 cm-1 is observed.
These changes in the CH2 stretching region clearly confirms the alteration in the
electron density of AHT amine group and, in effect, may indicate the development of
AHT:TNT Meisenheimer complex at pH 8.5 [4, 88, 176]. On the other hand, the
mixture of AHT and TNT at pH 7.0 does not show any significant change in the CH2
stretching region of AHT and thus indicate the lack of Meisenheimer complex at this
pH. Thus, both UV-Visible and solution Raman studies clearly indicate the
Meisenheimer complex between AHT and TNT at pH 8.5.
104 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Figure 4.2: (A) Solution Raman spectrum of 1 mM AHT at pH 8.5 (black), TNT at
pH 8.5 (blue), AHT:TNT mixture at pH 7 (green) and pH 8.5 (red). (B) The same
spectra as shown in A focused at the 2800-3000 cm-1 region.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 105
4.6.2 Electrochemical and XPS characterization of CA and AHT modified
electrodes/substrates
The next part of this study is to confirm the Meisenheimer complex between TNT
and AHT monolayer on pAu/AuNS surface. As we discussed in the introduction, the
upright orientation of AHT with free amine groups protruding outwards is critical for
the efficient complex formation between TNT and AHT. To confirm the orientation
of AHT SAM and the availability of free amine groups on the substrate’s surface,
after functionalization with AHT, electrochemical desorption and XPS experiments
were carried out. For comparison purposes, we also carried out electrochemical
desorption and XPS studies of pAu/AuNS surface that was modified with cysteamine
SAM. Figure 4.3 shows the cathodic linear sweep voltammograms (LSVs),
corresponding to desorption of cysteamine and AHT SAM from pAu/AuNS
electrode surface in 0.1 M KOH. The LSV of AHT (Figure 4.3, curve b) showed a
sharp cathodic peak at -1.0 V corresponding to the cleavage of Au-S bond [231]. The
sharp desorption peak of AHT clearly depicts that AHT has formed a well-packed
monolayer on the pAu/AuNS surface [231]. This observation provides indirect
evidence that all the AHT molecules are in the preferred trans conformation of
alkanethiols on the surface [231, 235, 238]. To the contrary, the desorption LSV of
cysteamine SAM from the pAu/AuNS surface shows two cathodic peaks at -0.91 V
and -0.77 V. The presence of two desorption peaks of the cysteamine monolayer may
be attributed in part to different Au-S bond environments and different orientations
of cysteamine on the pAu/AuS surface [92, 175, 239]. One possible orientation is the
vertical standing of cysteamine molecule where thiol is anchored to the Au surface
with the amine group pointing vertically away from the Au surface [92]. The other
possible orientation is the bridge-like orientation where the amine group of
cysteamine may also be bound to the Au surface in addition to the attachment of its
thiol group to the same surface [92]. Further, the desorption potential of Au-S bond
in the case of AHT monolayer is higher by ~ 0.1 V than that of the cysteamine
monolayer. This can be attributed to the higher stability of Au-S bond in the case of
AHT monolayer due to increased van der Waals lateral integration upon increasing
the hydrocarbon chain length of the alkyl thiol [231, 238]. Thus, the higher stability
of the AHT recognition molecule on the sensor surface is essential for the better shelf
life of the sensor.
106 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Figure 4.3: Linear sweep voltammograms obtained for the self-assembled
monolayer of (a) cysteamine (b) AHT in 0.1 M KOH at a scan rate of 0.1 Vs-1.
In order to validate the conclusions derived from the above electrochemical study we
carried out XPS investigation of pAu/AuNS surfaces that were modified with
cysteamine and AHT respectively. We examined the N 1s region of both cysteamine
and AHT modified surfaces. Figure 4.4 shows the raw and peak fitted spectra of the
N 1s region of cysteamine and AHT modified pAu/AuNS surfaces. The N 1s region
of AHT monolayer depicted a sharp peak at 400.1 eV (spectrum b in Figure 4.4)
confirming the existence of free amino groups on the substrate surface [231, 240].
On the other hand, the N 1s region of cysteamine monolayer showed two peaks at
400.1 eV and 398.3 eV. The peak at 400.1 eV is similar to that of the AHT
monolayer and corresponds to the free amine groups of cysteamine [231, 240]. The
second peak at the lower binding energy of 398.3 eV is in agreement with the
literature values for bound amine groups and, therefore, corresponds to the surface-
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 107
bound amine groups of cysteamine [231, 240, 241]. Further, cysteamine monolayer
also depicted higher intensity for the 398.3 eV peak in contrast to the 400.1 eV peak.
This may be attributed to the higher population of the “bridged” cysteamine
orientation over the vertical orientation confirming that majority of the amine groups
in cysteamine monolayer are not free. Thus, based on the electrochemical and XPS
study we conclude that the majority of the amine groups in AHT monolayer are
freely available on the pAu/AuNS surface. Therefore AHT represents a better
recognition layer for the SERS detection of TNT over cysteamine. In the
forthcoming studies, we employed pAu/AuNS surface that is modified with AHT
monolayer for the SERS based detection of TNT.
Figure 4.4: XPS spectra corresponding to N 1s region of self-assembled monolayer
of (a) cysteamine (b) AHT.
108 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
4.6.3 Characterization of AHT-TNT Meisenheimer complex on pAu/AuNS
surface
Figure 4.5A shows the SERS spectrum of AHT SAM (black spectrum) and
subsequent AHT:TNT complex (red spectrum) formed directly on the pAu/AuNS
surface. The SERS spectrum AHT SAM shows two CH2 stretching bands at 2855
cm-1 and 2919 cm-1. The SERS spectrum of AHT:TNT modified pAu/AuNS showed
intense signals at 1375 cm-1, 1559 cm-1 and 1608 cm-1. The high intensity of the
SERS signal at 1608 cm-1 (aromatic ring stretching) may be attributed in part to the
vertical orientation of the TNT:AHT complex over the substrate and the large Raman
cross-section of the aromatic ring when it is in a perpendicular position to the noble
metal surface [4, 103, 207, 208]. The spectrum also depicts a sharp and intense band
at 1375 cm-1 corresponding to NO2 group symmetric stretching vibration of TNT.
The band at 1559 cm-1 corresponds to asymmetric NO2 stretching of TNT [4, 103,
207, 208].
A closer examination of the AHT:TNT complex spectrum in the CH2 region (Figure
4.5A inset) shows two bands at 2855 cm-1 and 2919 cm-1 which are similar to those
of the AHT SAM. This indicates that there may be an abundance of free amine
groups on the substrate surface, which are not involved, in complex formation with
TNT. However, AHT forms dense monolayer on the surface and therefore TNT
molecules may experience some steric hindrance to reach the amine groups of the
AHT layer [242]. Therefore, optimizing the surface distribution of AHT is vital for
the effective complex formation between AHT and TNT. Beforehand, it is essential
to optimize the complex formation time between AHT and TNT on pAu/AuNS
surface. SERS measurements were carried out for the AHT-TNT complex by varying
the complex formation time (2 - 12 min) on the pAu/AuNS surface. Figure 4.5B
shows the corresponding time vs SERS intensity plot where 1608 cm-1 Raman band
of TNT was used as a reference band for quantifying the SERS intensity. Similar to
the solution phase AHT-TNT complex formation study, the complex formation on
surface (Figure 4.5B) also gets saturated around 10 min. Therefore, in all the
forthcoming studies, 10 min incubation time were permitted for the AHT-TNT
complex formation on pAu/AuNS surface.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 109
Figure 4.5: (A) SERS spectra of AHT monolayer (black) and AHT+TNT surface
bound complex on pAu/AuNS surface at pH 8.5. 1 nM TNT concentration was used
for complex formation. Inset shows the zoomed in amine stretching region. (B)
SERS intensity vs time plot for the formation of Meisenheimer complex between
AHT and TNT at pH 8.5.
110 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
4.6.4 AHT and BT mixed monolayer system for efficient TNT complex
formation
Mixed monolayer system is one of the best approaches to optimize the distribution of
the AHT molecules on the substrate surface and minimize the potential steric
hindrance of the TNT molecules attempting to bind the amine groups of the AHT
layer. In this study, we have chosen butanethiol (BT) as a spacer molecule for
developing a mixed monolayer of BT:AHT. BT is shorter than the AHT molecule by
two carbons. Therefore it can be used as a spacer between AHT molecules without
causing a steric hindrance to their amine groups. Further, the monolayer of BT
molecule has free methyl group at the terminal end while its thiol end is bound to the
Au surface. The free methyl groups do not have the ability to form a donor-acceptor
complex with TNT due to the lack of free lone pair of electrons when compared to
the amine groups on AHT. This is to say that, the methyl groups are not capable of
forming complex with the electron deficient TNT molecules. We have tested the
complex formation between BT and TNT at pH 8.5 using UV-Visible in solution
phase and also using SERS on pAu/AuNS surface. As expected, both showed
negative results confirming that BT is not able to form Meisenheimer complex with
TNT. Now, to arrive to the best ratio of the mixed monolayer system, we carried out
SERS measurements of TNT using Au/pAuNS surfaces modified with AHT/BT
mixed layers at various ratios of AHT to BT. Figure 4.6 shows the SERS spectrum of
TNT captured by pure AHT monolayer (black), AHT:BT (1:1) (blue) and AHT:BT
(9:1) (red) mixed layers. Among these three ratios, AHT:BT (9:1) mixed layer
showed the best TNT signal intensity and therefore we used it in the forthcoming
studies. Thus, the higher SERS intensity of TNT on AHT:BT (9:1) mixed layer
system clearly confirms that the shorter BT molecule served as an effective spacer
between AHT molecules at this 9:1 ratio. This provided easy access to large numbers
of TNT to bind to the pAu/AuNS surface via the AHT bridge in an almost
perpendicular orientation to the SERS substrate [175, 179, 200, 235] and
subsequently provide higher SERS signal. Further, the enhanced SERS intensity of
TNT on AHT:BT (9:1) mixed layer system indirectly conveys that on pure AHT
layer, the recognition molecules are closely packed and hence showed hindrance to
the effective complex formation with incoming TNT molecule.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 111
Figure 4.6: SERS spectra of TNT complex on AHT monolayer (black), AHT and
BT mixed monolayer in the ratio of 1:1 (blue) and 9:1 (red).
For the SERS quantification of TNT by present AHT:BT (9:1) modified pAu/AuNS
surface, various concentrations of TNT has been used to form Meisenheimer
complex over the sensor surface. Figure 4.7A shows the SERS spectra of TNT
between 1 nM and 100 fM. The band at 1608 cm-1 was used as a reference band for
the quantitative analysis of TNT. The intensity of TNT SERS signal increased
monotonically with the increase in TNT concentration. Figure 4.7B illustrates the
linear correlation between the logarithmic SERS signal intensity at 1608 cm-1 and the
corresponding logarithmic solution concentration of TNT [97]. A good linear
response was achieved within the given concentration range of 1 nM to 100 fM of
TNT with a correlation coefficient of 0.9910. The error bars indicate the standard
deviations from 10 measurements at different spots over the SERS substrate. The
lower limit of quantification (LQD) for TNT by the present sensor was found to be
100 fM. The theoretical limit of detection (LOD) can be obtained from the x-axis
intercept of linear regression [97, 138]. We estimated the theoretical LOD to be 6.3 ×
10-14 M for the present TNT sensor [97].
112 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Figure 4.7: (A) SERS spectra @1608 cm-1 obtained for various concentrations of
TNT on AHT:BHT (9:1) modified pAu/AuNS surface at pH 8.5. (B) Plot
demonstrating the linear correlation between logarithmic solution concentration of
TNT and logarithmic SERS intensity of the AHT-TNT complex.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 113
4.6.5 Interference study
2,4-DNT and PA are the two major contaminants that usually co-exist with TNT
[200]. Further, both have Raman fingerprints similar to TNT. Therefore, it is very
important to selectively identify TNT in presence of DNT and PA in order to
demonstrate the method for practical in-field applications. We reacted both 2,4-DNT
and PA individually with AHT at pH 8.5 and subsequently monitored the UV-Visible
spectrum. The UV-Vis spectra of the 2,4-DNT/AHT and PA/AHT mixtures did not
show any change to the absorption spectra of 2,4-DNT or PA (Figure 4.8). Also, the
visual inspection did not indicate any color change unlike the AHT-TNT reaction.
This clearly indicates that AHT selectively forms a Meisenheimer complex with
TNT while completely disregarding 2,4-DNT and PA even though both molecules
are also electron deficient. The reason for this is the lack of nitro group in 2,4-DNT
prevents distribution of anionic charge throughout the benzene ring and therefore
2,4-DNT may not form Meisenheimer complex with AHT [4]. Although PA has a
similar structure to TNT, the –OH group can easily lose a proton which would reduce
the charge transfer between AHT and PA and thus less likely to form a complex
[207]. In order to further confirm the selectivity of the AHT towards TNT, SERS
investigation of 2,4-DNT and PA was carried out on AHT modified surface. Both
AHT-DNT and AHT-PA modified surfaces were prepared similar to that of AHT-
TNT modified surface as presented in the experimental section. As expected, the
SERS spectra of AHT-DNT and AHT-PA surface do not show Raman bands
corresponding to aromatic NO2 stretching except the SERS spectral features similar
to AHT monolayer (Figure 4.2A black spectrum). This clearly confirms that AHT is
does not form Meisenheimer complex with 2,4-DNT and PA on surface and hence
no interference.
114 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Figure 4.8: UV-Vis absorption spectra of 2,4-DNT (green), PA (brown), TNT
(blue), AHT+DNT (black), AHT+PA (red) and AHT+TNT (pink) at pH 8.5.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 115
4.6.6 Detection of TNT in soil
In order to demonstrate the practical application of the present sensor, we carried out
the detection of TNT spiked into soil. Before spiking the TNT into soil, the soil
sample was tested for any pre-existing TNT. The SERS spectrum of the soil extract,
before spiking with TNT, confirmed the absence of pre-existing TNT. 2 nM TNT
was then spiked in soil and subsequently extracted into the aqueous medium. The
TNT extract was then screened by pAu/AuNS SERS substrate modified with 9:1
AHT:BT mixed monolayer (Figure 4.9). The SERS spectrum of extracted TNT
showed similar fingerprint pattern to that of TNT standard. To demonstrate the
repeatability of TNT detection from soil, SERS measurements were repeated on two
separate dates with 10 SERS spectra measured on each date (total n = 20). The
reproducibility of the SERS sensor was determined based on the SERS signal
obtained from random spots across the entire substrate under the same experimental
conditions. The relative standard deviation of the characteristic SERS peaks of TNT
for the 20 measurements was found to be 5.23% confirming the reproducibility of the
sensor. The concentration of TNT extracted by water medium was calculated from
the calibration curve (Figure 4.7B) and was found to be 302 pM only. This is due to
the poor extraction of TNT that is caused by its poor solubility in water (0.13 g/L at
20 °C) [243]. Despite the low recovery, these results clearly demonstrate that the
present methodology can be employed to identify TNT in field for environmental
and forensic applications.
116 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Figure 4.9: SERS spectra of 1 nM (black) and 2 nM soil extracted (red) TNT on
AHT:BHT (9:1) modified pAu/AuNS surface at pH 8.5.
Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment 117
4.7 CONCLUSION
In this article, we have demonstrated a simple, rapid and selective approach for the
detection of trace amounts of TNT in aqueous media via SERS. The method utilized
AHT as the new TNT recognition molecule. The Meisenheimer complex formed
between AHT and TNT was supported by UV-Visible and Raman spectroscopy. The
complex formation was rapid and the reaction completed in almost 10 min, which
laid platform for the rapidity of the present method. Prior to the formation of
AHT:TNT complex on the surface, AHT SAM modified surface was characterized
by electrochemical desorption and XPS studies. Both the studies provided supportive
evidence for the availability of free amine groups on the surface with suitable
orientation. AHT SAM and AHT:BT mixed SAM system was explored to enhance
the TNT SERS signal. AHT:BT mixed SAM in 9:1 ratio delivered higher TNT SERS
signal with a LOD of 100 fM. Further, the new sensor is selective towards TNT over
2,4-DNT and PA. Thus, the present method is rapid and selective which satisfies the
requirements for the detection of TNT in real life samples. Finally, this new sensor
successfully demonstrated the detection of TNT in soil with a calibration curve
developed between 1 nM and 100 fM of TNT concentration. Thus, the present TNT
sensor can be used for the rapid screening of TNT residues in soil and water.
4.8 ACKNOWLEDGEMENT
The Authors sincerely thank Bharati Gupta, CPME, QUT for XPS analysis and
discussions.
118 Aminohexanethiol mixed monolayer nanosensor for SERS detection of TNT in the environment
Conclusions 119
Chapter 5: Conclusions
Tailoring the surface chemistry of SERS substrates to create specificity holds
considerable promise for improving explosives detection. More importantly, this
methodology could open the gateway for further development and application in the
industry as an alternative to existing methods for trace detection of explosives. The
main findings derived from this study demonstrated a simple, practical in-field
device for detection of energetic materials. Building upon the strengths and
limitations of current methods led to an improved rapid, reliable, sensitive and
selective SERS detection system. The three different methodologies introduced
throughout this dissertation were successfully utilized for detection of TNT at trace
levels. Our interest was in the development of suitable assemblies onto the surface of
gold substrates for trace detection of TNT. The strategy for forming functionalized
assemblies on metal substrate surface involved aminothiols as linkers to bind to gold
on the thiol end and capture the target molecule with its amine moiety
simultaneously. The efficiency of these techniques was determined based on their
sensitivity and selectivity towards TNT by investigating the signal intensity of the
resulting spectra. The underlying principle to balance between both sensitivity and
selectivity was the formation of Meisenheimer complex, which is widely used for
TNT detection. In this work, the applicability of two aminothiols, specifically
cysteamine and 6-aminohexanethiol as recognition molecules for SERS detection of
TNT in aqueous solution and environment was successfully demonstrated. In
contrast to other organic molecules, aminothiols were selected because of its high
affinity for gold, eliminating the need for complex preparations for assembly onto the
metal surface. The fabrication of a more uniform and reproducible metal substrate
was vital to the success of this research. The practical significance gained from the
approach used includes:
1- ease of sample collection without laborious pre-treatment;
2- reproducibility in SERS signal;
3- selectivity over possible coexistent interfering molecules;
4- high sensitivity with low limit of detection;
5- field deployable and can be adaptable to hand held devices.
120 Conclusions
Initial work using gold colloids functionalized with cysteamine was found to be
sensitive. As previous studies have shown that cysteamine can adopt two different
conformations, it was important to tailor the experiment to ensure the amine end
group binds to TNT instead of to the gold surface. The strategy used was to form the
Meisenheimer complex between cysteamine and TNT prior to introducing the
complex to the gold surface. This way, only the thiol end was free to bind to gold.
The formation of Meisenheimer complex between cysteamine and TNT allowed
SERS detection down to the concentration level of 22.7 ng/L, which is lower than the
minimum acceptable limit of 0.04 mg/L for TNT in wastewater. Visual color change,
UV-Vis and SERS data confirmed the formation of Meisenheimer complex with
TNT over two closely related molecules, 2,4-DNT and PA. The interference study
revealed lack of signal in the presence of 2,4-DNT and PA, implying its selectivity
for TNT. However, the developed colloidal SERS method for TNT detection tends to
be less reproducible in terms of both fabrication and signal response. The instability
from uncontrolled aggregation of nanoparticles may also lead to irreversible
aggregation rendering them unsuitable for routine analysis.
After understanding the methodology of cysteamine assembly onto the gold surface,
we explored the potential of different SERS substrates by tuning their SERS
properties. To utilize the potential of the Meisenheimer complex whilst overcoming
the challenging limitation of colloidal gold, fabrication of nanostructured gold
substrates was considered. The same mechanism of forming the Meisenheimer
complex between cysteamine and TNT first, was employed. Our data indicated that
the closed packed morphology of the nanostructured substrate delivered much better
reproducibility and enhancement of the acquired TNT signal. A lower RSD of 4.37%
was achieved when combined with WAI. High sensitivity down to femtogram levels
were achieved through the Meisenheimer complex of cysteamine-TNT assembled
onto the substrate. The capability to control the properties of the nanostructured
substrate contributed to the higher SERS activity over colloidal gold. Additionally,
the nanostructured substrate surfaces have shown suitability for in-field use, as well
as selectivity over 2,4-DNT and PA. The nanostructured SERS substrate was
successfully utilized for the detection of TNT in aqueous solutions and soil extracts
without the need for any tedious sample preparation or pre-concentration. The
Conclusions 121
nanostructured substrate performed significantly better than colloidal gold
nanoparticles. The present work also suggests different adsorption configurations of
TNT when absorbed onto nanoparticles in comparison to nanostructured surfaces.
Gold nanoparticles depicted enhanced NO2 symmetric stretching as the prominent
band whereas the nanostructured substrate enhanced the C=C aromatic stretching
vibration the most.
The slight downside in which cysteamine tends to bend backwards for the amine end
to attach to the metal surface has prompted additional improvements for a more
suitable chemical moiety that could bring about better enhancements and easier
control over molecule absorption onto the metal. Our strategy was to use a longer
chained aminothiol that is known to assemble in an upward direction but not too long
as to push away the target molecule further from the metal surface. AHT was used as
a new moiety to target TNT. To the best of our knowledge, AHT has not been
utilized for SERS detection of TNT. The six-carbon chain of AHT ensures that it
assembles in an almost vertical position leaving the amine group free and ready to
react with TNT. This was reiterated by the XPS data obtained confirming the higher
abundance of free amine groups with AHT compared to cysteamine. However, the
closely packed assembly of AHT caused a limited amount of detectable TNT
possibly due to less accessible binding sites. To address this limitation, a mixed
monolayer of AHT with BT was evaluated. The use of BT created adequate spacing
for TNT to position itself in a vertical orientation onto the amine moiety of AHT. In
addition, the ability of AHT to form a mixed monolayer with BT as gap fillers
increased the selectivity of the system by preventing smaller molecules from binding
to the metal substrate. This has intensified the sensitivity of the nanostructured
substrates when functionalized with AHT as the recognition molecule. High
sensitivity, selectivity and reproducibility with lack of interference from PA and 2,4-
DNT were established through both Raman and UV-Vis studies. The strength and
versatility of this method were tested under lab conditions as well as with
environmental samples to ensure reliable detection for both lab operations and under
field conditions.
122 Conclusions
The assembly of nanostructured gold substrates brought about a low detection limit
of 100 fM regardless whether cysteamine or AHT was used. Extraction of TNT from
soil samples also showed high sensitivity with comparable Raman signal obtained at
1 nM. One key advantage of the efficiency of AHT as a recognition molecule over
cysteamine is the ability to readily form highly ordered SAMs on a solid substrate
and in effect delivers a functionalized SERS substrate suitable for in-field
applications. In contrast, cysteamine as a recognition molecule must be used in
solution form prior to loading the cysteamine-TNT complex onto the metal substrate
in order to avoid potential inconsistent measurements from the different orientations
of cysteamine on surfaces. This illustrates the importance of selecting an appropriate
aminothiol for SERS detection. The many advantages of AHT over cysteamine give
AHT an edge as a desirable recognition molecule for trace detection of TNT.
Overall, the objectives of this research were met with the fabrication of a reliable
sensing platform for SERS. The versatility of the fabricated nanosensor provided an
effective detection with reproducibility, high selectivity and high sensitivity needed
in routine trace explosive analysis. We hope to have contributed to the potential use
of functionalized nanostructured substrates as a highly attractive technique for
identification and characterization of explosive residues in minute amounts.
5.1 FUTURE WORK
This simple sensing system of functionalized nanosensors for detecting energetic
materials based on SERS paves the way for future improved experiments. This work
is a first step towards the goal of better detection systems for trace detection. A few
key elements can be addressed before the nanosensor can be usable and practical for
real-world implementation. For optimization of the nanostructured substrate,
parameters such as different electrodeposition time, increased number of potential
scans, different electrolyte temperature and pre-treatment of pAu/AuNS substrate can
be trialled to tailor the morphology of the substrate for maximum SERS
enhancement. Obtaining the appropriate optimal parameters would ensure better
measurement efficiency. For a more cost-effective approach, the replacement of
AuNPs with CuNPs, which has a similar SPR to AuNPs can be considered.
Conclusions 123
Substituting Au with silicon, graphene or other non-precious metals as the substrate
material may serve as a more cost-effective detection system for explosive materials.
It is also worthwhile to explore the reusability of the nanostructured surface to reduce
the need for fabrication of new surface for every analysis. It would be interesting to
experiment the efficiency rate of the SERS substrate after every use and wash cycle
and consequently determine its reusability rate. In addition, continuing work is
required to further develop new selective SERS sensors for a variety of explosives in
field. The developed strategy can be easily extended to the screening and
characterization of other different energetic materials provided that a suitable
recognition molecule is obtained for the target analyte.
Another future research direction is to apply this methodology for vapor phase
detection. Explosives detection in vapor phase still remains difficult because of the
low vapor pressure of many of the energetic materials. The nanostructured substrate
has demonstrated unique sensitivity down to 100 fM level offering advantages of
reliable detection of low-level energetic materials. This limit of detection is below
the 1 ppb level benchmark for detection of nitroaromatic vapours for air quality
monitoring [28]. A possible extension to the presented work is to introduce
microfluidic devices for gas phase detection of TNT with a SERS detection mode.
The microfluidic device would incorporate a nebulizer unit that bubbles the gas
phase sample into a basic solution of amino thiols (cysteamine or AHT) to form a
Meisenheimer complex within the nano channels of the device. The formed complex
will flow to the detection section where it will bind to the surface of the
nanostructured substrate via the thiol moiety of the formed Meisenheimer complex.
The nanostructured substrate will act as a SERS sensor for the direct detection of the
captured TNT. Memory effects of previous samples can be eliminated through wash
cycles of piranha solution after each detection cycle. Coupling the substrate with a
portable miniature microfluidic device will develop a portable Raman detection
system capable of detecting TNT in gas phase at ultra-trace levels.
It is also envisioned that the SERS nanosensor presented in this study be extended to
disposable strips to ease application in field. Fabrication of disposable strips based on
the concept of the AHT assemblies on nanostructured substrates will provide a
simple cost-effective portable platform for SERS detection of TNT. The nanosensor
124 Conclusions
strips would provide a low-cost, easy-to-use, rapid on-site detection of explosive
samples. For quantitative real-time detection, the strips can be easily interfaced with
handheld Raman devices, such as the one illustrated in Figure 5.1, for output signals.
We believe that with these additional research works, the SERS substrates can be
implemented in a portable device for real-time detection, as well as contribute to a
wider variety of applications.
Figure 5.1: A handheld Raman device that can easily be interfaced with other
platforms. Adapted from reference [226].
References 125
References
1. Guo, Z., et al., Ultrasensitive trace analysis for 2,4,6-trinitrotoluene using
nano-dumbbell surface-enhanced Raman scattering hot spots. Analyst, 2014.
139(4): p. 807-812.
2. Izake, E.L., Forensic and homeland security applications of modern portable
Raman spectroscopy. Forensic Science International, 2010. 202(1–3): p. 1-8.
3. Caygill, J.S., F. Davis, and S.P.J. Higson, Current trends in explosive
detection techniques. Talanta, 2012. 88: p. 14-29.
4. Dasary, S.S.R., et al., Gold nanoparticle based label-free SERS probe for
ultrasensitive and selective detection of trinitrotoluene. Journal of the
American Chemical Society, 2009. 131(38): p. 13806-13812.
5. Almaviva, S., et al., Application of micro-Raman spectroscopy for fight
against terrorism and smuggling. Optical Engineering, 2014. 53(4): p.
044113-044113.
6. Muro, C.K., et al., Vibrational Spectroscopy: Recent Developments to
Revolutionize Forensic Science. Analytical Chemistry, 2015. 87(1): p. 306-
327.
7. Chen, Y., et al., L-cysteine-capped CdTe QD-based sensor for simple and
selective detection of trinitrotoluene. Nanotechnology, 2010. 21(12): p.
125502.
8. Singh, S., Sensors—an effective approach for the detection of explosives.
Journal of Hazardous Materials, 2007. 144(1): p. 15-28.
9. Ma, Y., S. Wang, and L. Wang, Nanomaterials for luminescence detection of
nitroaromatic explosives. TrAC Trends in Analytical Chemistry, 2015. 65: p.
13-21.
10. Germain, M.E. and M.J. Knapp, Optical explosives detection: from color
changes to fluorescence turn-on. Chemical Society Reviews, 2009. 38(9): p.
2543-2555.
11. Harper, R.J. and M.E. Fisher, Intricacies of Comparative Testing of
Explosives Detectors at the Ultra-Trace Level, in Chemical, Biological,
126 References
Radiological, Nuclear, and Explosives, A.W. Fountain Iii and P.J. Gardner,
Editors. 2010.
12. Golightly, R.S., W.E. Doering, and M.J. Natan, Surface-Enhanced Raman
Spectroscopy and Homeland Security: A Perfect Match? Acs Nano, 2009.
3(10): p. 2859-2869.
13. Ryder, A.G., Surface enhanced Raman scattering for narcotic detection and
applications to chemical biology. Current Opinion in Chemical Biology,
2005. 9(5): p. 489-493.
14. Massarini, E., et al., Methodologies for assessment of limit of detection and
limit of identification using surface-enhanced Raman spectroscopy. Sensors
and Actuators B: Chemical, 2015. 207: p. 437-446.
15. Lin, D.Y., et al., Ultrasensitive optical detection of trinitrotoluene by
ethylenediamine-capped gold nanoparticles. Analytica Chimica Acta, 2012.
744: p. 92-98.
16. Moore, D.S., Instrumentation for trace detection of high explosives. Review
of Scientific Instruments, 2004. 75(8): p. 2499-2512.
17. Yinon, J., Detection of Explosives by Electronic Noses. Analytical Chemistry,
2003. 75(5): p. 98 A-105 A.
18. Otto, J., M.F. Brown, and W. Long Iii, Training rats to search and alert on
contraband odors. Applied Animal Behaviour Science, 2002. 77(3): p. 217-
232.
19. Corcelli, A., et al., Detection of explosives by olfactory sensory neurons.
Journal of hazardous materials, 2010. 175(1): p. 1096-1100.
20. Shaw, J., et al., Polarization lidar measurements of honey bees in flight for
locating land mines. Optics Express, 2005. 13(15): p. 5853-5863.
21. Dickinson, T.A., et al., Convergent, Self-Encoded Bead Sensor Arrays in the
Design of an Artificial Nose. Analytical Chemistry, 1999. 71(11): p. 2192-
2198.
22. Albert, K.J., et al., Field-Deployable Sniffer for 2,4-Dinitrotoluene Detection.
Environmental Science & Technology, 2001. 35(15): p. 3193-3200.
23. Ewing, R.G., et al., A critical review of ion mobility spectrometry for the
detection of explosives and explosive related compounds. Talanta, 2001.
54(3): p. 515-529.
References 127
24. Hill, H.H. and G. Simpson, Capabilities and limitations of ion mobility
spectrometry for field screening applications. Field Analytical Chemistry &
Technology, 1997. 1(3): p. 119-134.
25. Yang, X., et al., Fluorescent detection of TNT and 4-nitrophenol by BSA Au
nanoclusters. Dalton Transactions, 2014. 43(26): p. 10057-10063.
26. McIntosh, A.I., et al., Terahertz spectroscopy: a powerful new tool for the
chemical sciences? Chemical Society Reviews, 2012. 41(6): p. 2072-2082.
27. Federici, J.F., et al., THz imaging and sensing for security applications—
explosives, weapons and drugs. Semiconductor Science and Technology,
2005. 20(7): p. S266.
28. Chou, A., et al., SERS substrate for detection of explosives. Nanoscale, 2012.
4(23): p. 7419-7424.
29. Steinfeld, J.I. and J. Wormhoudt, Explosives detection: a challenge for
physical chemistry. Annual Review of Physical Chemistry, 1998. 49(1): p.
203-232.
30. Adar, F., Molecular Spectroscopy Workbench: Raman Spectroscopy for
Biomedical Applications, 2009, Advanstar Communications, Inc. p. 26.
31. Kneipp, K., et al., Ultrasensitive Chemical Analysis by Raman Spectroscopy.
Chemical Reviews, 1999. 99(10): p. 2957-2976.
32. Smith, W.E., Practical understanding and use of surface enhanced Raman
scattering/surface enhanced resonance Raman scattering in chemical and
biological analysis. Chemical Society Reviews, 2008. 37(5): p. 955-964.
33. Le Ru, E. and P. Etchegoin, Principles of Surface-Enhanced Raman
Spectroscopy: and related plasmonic effects. 2008: Elsevier.
34. Wang, X.F., et al., Detection of TNT in acetone using Raman spectroscopic
signature - art. no. 662219, in International Symposium on Photoelectronic
Detection and Imaging 2007: Laser, Ultraviolet, and Terahertz Technology,
L. Zhou, Editor. 2008. p. 62219-62219.
35. Wackerbarth, H., et al., Detection of explosives based on surface-enhanced
Raman spectroscopy. Applied Optics, 2010. 49(23): p. 4362-4366.
36. Fleischmann, M., P.J. Hendra, and A.J. McQuillan, Raman spectra of
pyridine adsorbed at a silver electrode. Chemical Physics Letters, 1974.
26(2): p. 163-166.
128 References
37. Jeanmaire, D.L. and R.P. Van Duyne, Surface Raman
spectroelectrochemistry: Part I. Heterocyclic, aromatic, and aliphatic amines
adsorbed on the anodized silver electrode. Journal of Electroanalytical
Chemistry and Interfacial Electrochemistry, 1977. 84(1): p. 1-20.
38. Albrecht, M.G. and J.A. Creighton, Anomalously intense Raman spectra of
pyridine at a silver electrode. Journal of the American Chemical Society,
1977. 99(15): p. 5215-5217.
39. Camden, J.P., et al., Probing the structure of single-molecule surface-
enhanced Raman scattering hot spots. Journal of the American Chemical
Society, 2008. 130(38): p. 12616-12617.
40. Otto, A., What is observed in single molecule SERS, and why? Journal of
Raman Spectroscopy, 2002. 33(8): p. 593-598.
41. Nie, S. and S.R. Emory, Probing Single Molecules and Single Nanoparticles
by Surface-Enhanced Raman Scattering. Science, 1997. 275(5303): p. 1102-
1106.
42. Doering, W.E. and S. Nie, Single-Molecule and Single-Nanoparticle SERS:
Examining the Roles of Surface Active Sites and Chemical Enhancement. The
Journal of Physical Chemistry B, 2001. 106(2): p. 311-317.
43. Tian, Z.-Q., et al., Expanding generality of surface-enhanced Raman
spectroscopy with borrowing SERS activity strategy. Chemical
Communications, 2007(34): p. 3514-3534.
44. Halvorson, R.A. and P.J. Vikesland, Surface-Enhanced Raman Spectroscopy
(SERS) for Environmental Analyses. Environmental Science & Technology,
2010. 44(20): p. 7749-7755.
45. Lis, D. and F. Cecchet, Localized surface plasmon resonances in
nanostructures to enhance nonlinear vibrational spectroscopies: towards an
astonishing molecular sensitivity. Beilstein journal of nanotechnology, 2014.
5(1): p. 2275-2292.
46. McCreery, R.L., Raman spectroscopy for chemical analysis. Vol. 225. 2005:
John Wiley & Sons.
47. Aroca, R., Surface-enhanced vibrational spectroscopy. 2006: John Wiley &
Sons.
48. Cyrankiewicz, M., T. Wybranowski, and S. Kruszewski. Study of SERS
efficiency of metallic colloidal systems. IOP Publishing.
References 129
49. Botti, S., et al., Assessment of SERS activity and enhancement factors for
highly sensitive gold coated substrates probed with explosive molecules.
Chemical Physics Letters, 2014. 592(0): p. 277-281.
50. Smith, E. and G. Dent, Modern Raman spectroscopy: a practical approach.
2013: John Wiley & Sons.
51. Ko, H., S. Singamaneni, and V.V. Tsukruk, Nanostructured surfaces and
assemblies as SERS media. Small (Weinheim an der Bergstrasse, Germany),
2008. 4(10): p. 1576-1599.
52. Tognalli, N.G., et al., Exploring three-dimensional nanosystems with Raman
spectroscopy: methylene blue adsorbed on thiol and sulfur monolayers on
gold. The Journal of Physical Chemistry B, 2006. 110(1): p. 354-360.
53. Otto, A., The ‘chemical’(electronic) contribution to surface‐ enhanced
Raman scattering. Journal of Raman Spectroscopy, 2005. 36(6‐ 7): p. 497-
509.
54. Jiang, K., I. White, and D.L. DeVoe. Detection of trace explosives by SERS
using 3-D nanochannel arrays.
55. Freus, G., Controlled nucleation for the regulation of the particle size in
monodisperse gold solutions. Nature (Phys Sci), 1973. 241: p. 20-22.
56. Bengter, H., C. Tengroth, and S.P. Jacobsson, New light on Ag‐ colloid
preparation for surface‐ enhanced FT‐ Raman spectroscopy: the role of
aggregation. Journal of Raman Spectroscopy, 2005. 36(11): p. 1015-1022.
57. Klutse, C.K., et al., Applications of self-assembled monolayers in surface-
enhanced Raman scattering. Journal of Nanotechnology, 2012. 2012.
58. Sivanesan, A., et al., Towards improved precision in the quantification of
surface-enhanced Raman scattering (SERS) enhancement factors: a renewed
approach. Analyst, 2015. 140(2): p. 489-496.
59. Kosuda, K.M., et al., 9 Nanostructures and Surface-Enhanced Raman
Spectroscopy. Handbook of Nanoscale Optics and Electronics, 2010: p. 309.
60. Love, J.C., et al., Self-assembled monolayers of thiolates on metals as a form
of nanotechnology. Chemical reviews, 2005. 105(4): p. 1103-1170.
61. Laibinis, P.E., et al., Comparison of the structures and wetting properties of
self-assembled monolayers of n-alkanethiols on the coinage metal surfaces,
copper, silver, and gold. Journal of the American Chemical Society, 1991.
113(19): p. 7152-7167.
130 References
62. Ulman, A., Formation and structure of self-assembled monolayers. Chemical
reviews, 1996. 96(4): p. 1533-1554.
63. Bharadwaj, R. and S. Mukherji, Gold nanoparticle coated U-bend fibre optic
probe for localized surface plasmon resonance based detection of explosive
vapours. Sensors and Actuators B-Chemical, 2014. 192: p. 804-811.
64. Dasary, S.S.R., et al., Ultrasensitive and highly selective detection of TNT
from environmental sample using two-photon scattering properties of
aminothiophenol-modified gold nanoparticle. IEEE Transactions on
Nanotechnology, 2011. 10(5): p. 1083-1088.
65. Xu, Z., et al., Surface-Enhanced Raman Scattering Spectroscopy of Explosive
2,4-Dinitroanisole using Modified Silver Nanoparticles. Langmuir, 2011.
27(22): p. 13773-13779.
66. Mohan, M. and D.K. Chand, Visual colorimetric detection of TNT and 2,4-
DNT using as-prepared hexaazamacrocycle-capped gold nanoparticles.
Analytical Methods, 2014. 6(1): p. 276-281.
67. Jiang, Y., et al., A Simple Assay for Direct Colorimetric Visualization of
Trinitrotoluene at Picomolar Levels Using Gold Nanoparticles. Angewandte
Chemie-International Edition, 2008. 47(45): p. 8601-8604.
68. Pandya, A., et al., A novel nanoaggregation detection technique of TNT using
selective and ultrasensitive nanocurcumin as a probe. Analyst, 2012. 137(8):
p. 1771-1774.
69. Glembocki, O.J., et al. Cysteamine coated Ag and Au nanorods for improved
surface enhanced Raman scattering from dinitrotoluene and trinitrotoluene.
in Proceedings of SPIE. 2010.
70. Bensebaa, F., et al., Kinetics of formation of long-chain n-alkanethiolate
monolayers on polycrystalline gold. Langmuir, 1997. 13(20): p. 5335-5340.
71. Bain, C.D., et al., Formation of monolayer films by the spontaneous assembly
of organic thiols from solution onto gold. Journal of the American Chemical
Society, 1989. 111(1): p. 321-335.
72. Love, J.C., et al., Formation and structure of self-assembled monolayers of
alkanethiolates on palladium. Journal of the American Chemical Society,
2003. 125(9): p. 2597-2609.
73. Kawasaki, M., et al., Rapid self-assembly of alkanethiol monolayers on
sputter-grown Au (111). Langmuir, 2000. 16(4): p. 1719-1728.
References 131
74. Yamada, R., H. Wano, and K. Uosaki, Effect of temperature on structure of
the self-assembled monolayer of decanethiol on Au (111) surface. Langmuir,
2000. 16(13): p. 5523-5525.
75. Bain, C.D., J. Evall, and G.M. Whitesides, Formation of monolayers by the
coadsorption of thiols on gold: variation in the head group, tail group, and
solvent. Journal of the American Chemical Society, 1989. 111(18): p. 7155-
7164.
76. Bain, C.D. and G.M. Whitesides, Formation of two-component surfaces by
the spontaneous assembly of monolayers on gold from solutions containing
mixtures of organic thiols. Journal of the american chemical society, 1988.
110(19): p. 6560-6561.
77. Solomons, T.W.G. and C. Fryhle, Organic Chemistry, John Wiley&Sons.
Inc., 6thed, 2000: p. 646-931.
78. Terrier, F., Rate and equilibrium studies in Jackson-Meisenheimer
complexes. Chemical Reviews, 1982. 82(2): p. 77-152.
79. De La Cruz-Montoya, E., et al., Enhanced Raman spectroscopy of 2,4,6-TNT
in anatase and rutile Titania nanocrystals - art. no. 62030X, in Optics and
Photonics in Global Homeland Security II, T.T. Saito and D. Lehrfeld,
Editors. 2006, Spie-Int Soc Optical Engineering: Bellingham. p. X2030-
X2030.
80. Demeritte, T., et al., Highly efficient SERS substrate for direct detection of
explosive TNT using popcorn-shaped gold nanoparticle-functionalized
SWCNT hybrid. Analyst, 2012. 137(21): p. 5041-5045.
81. Zhou, X., et al., SERS and OWGS detection of dynamic trapping molecular
TNT based on a functional self-assembly Au monolayer film. Analyst, 2013.
138(6): p. 1858-1864.
82. Qu, W.-G., et al., Plasmonic resonance energy transfer-based
nanospectroscopy for sensitive and selective detection of 2,4,6-trinitrotoluene
(TNT). Chemical Communications, 2011. 47(4): p. 1237-1239.
83. Feng, L., et al., Detection of TNT based on conjugated polymer encapsulated
in mesoporous silica nanoparticles through FRET. Chemical
Communications, 2012. 48(38): p. 4633-4635.
132 References
84. Qian, K., et al., Functionalized shell-isolated nanoparticle-enhanced Raman
spectroscopy for selective detection of trinitrotoluene. Analyst, 2012.
137(20): p. 4644-4646.
85. Chen, H., H. Chen, and R.G. Cooks, Meisenheimer complexes bonded at
carbon and at oxygen. Journal of the American Society for Mass
Spectrometry, 2004. 15(7): p. 998-1004.
86. Fant, F., et al., The use of amino compounds for binding 2,4,6-trinitrotoluene
in water. Environmental Pollution, 2001. 111(3): p. 503-507.
87. Hutchison, J.A., et al., Subdiffraction Limited, Remote Excitation of Surface
Enhanced Raman Scattering. Nano Letters, 2009. 9(3): p. 995-1001.
88. Riauba, L., et al., A Study of Cysteamine Ionization in Solution by Raman
Spectroscopy and Theoretical Modeling. The Journal of Physical Chemistry
A, 2006. 110(50): p. 13394-13404.
89. Jerez-Rozo, J.I., et al., Enhanced Raman Scattering of 2,4,6TNT Using
Metallic Colloids. IEEE Sensors Journal, 2008. 8(6): p. 974-982.
90. Kneipp, K., et al., Near-infrared surface-enhanced Raman scattering of
trinitrotoluene on colloidal gold and silver. Spectrochimica Acta Part A:
Molecular and Biomolecular Spectroscopy, 1995. 51(12): p. 2171-2175.
91. Michota, A., A. Kudelski, and J. Bukowska, Chemisorption of cysteamine on
silver studied by surface-enhanced Raman scattering. Langmuir, 2000.
16(26): p. 10236-10242.
92. Zhang, J., et al., Coexistence of multiple conformations in cysteamine
monolayers on Au (111). The Journal of Physical Chemistry B, 2005.
109(32): p. 15355-15367.
93. Mocanu, A., et al., Self-assembly characteristics of gold nanoparticles in the
presence of cysteine. Colloids and Surfaces A: Physicochemical And
Engineering Aspects, 2009. 338(1): p. 93-101.
94. Horovitz, O., et al., Protein Binding to Gold Colloids. Gold Bulletin, 2007.
40(3): p. 213-218.
95. Weng, Z., et al., Self-assembly of core-satellite gold nanoparticles for
colorimetric detection of copper ions. Analytica Chimica Acta, 2013. 803(0):
p. 128-134.
References 133
96. Koneswaran, M. and R. Narayanaswamy, L-Cysteine-capped ZnS quantum
dots based fluorescence sensor for Cu 2+ ion. Sensors and Actuators B:
Chemical, 2009. 139(1): p. 104-109.
97. Liu, M. and W. Chen, Graphene nanosheets-supported Ag nanoparticles for
ultrasensitive detection of TNT by surface-enhanced Raman spectroscopy.
Biosensors and Bioelectronics, 2013. 46: p. 68-73.
98. Yang, L., et al., Ultrasensitive SERS detection of TNT by imprinting
molecular recognition using a new type of stable substrate. Chemistry – A
European Journal, 2010. 16(42): p. 12683-12693.
99. Holthoff, E.L., D.N. Stratis-Cullum, and M.E. Hankus, A nanosensor for TNT
detection based on molecularly imprinted polymers and surface enhanced
Raman scattering. Sensors, 2011. 11(3): p. 2700-2714.
100. Mathew, A., P.R. Sajanlal, and T. Pradeep, Selective visual detection of TNT
at the sub-zeptomole level. Angewandte Chemie-International Edition, 2012.
51(38): p. 9596-9600.
101. Wackerbarth, H., et al., Challenge of false alarms in nitroaromatic explosive
detection-a detection device based on surface-enhanced Raman spectroscopy.
Applied Optics, 2010. 49(23): p. 4367-4371.
102. Lewis, I.R., N.W. Daniel, and P.R. Griffiths, Interpretation of Raman spectra
of nitro-containing explosive materials. Part I: Group frequency and
structural class membership. Applied spectroscopy, 1997. 51(12): p. 1854-
1867.
103. Kanchanapally, R., et al., Graphene Oxide–Gold Nanocage Hybrid Platform
for Trace Level Identification of Nitro Explosives Using a Raman
Fingerprint. The Journal of Physical Chemistry C, 2014. 118(13): p. 7070-
7075.
104. Spencer, K.M., et al., Advances in landmine detection using surface-
enhanced Raman spectroscopy, in Detection and Remediation Technologies
for Mines and Minelike Targets Iv, Pts 1 and 2, A.C. Dubey, et al., Editors.
1999, Spie-Int Soc Optical Engineering: Bellingham. p. 373-379.
105. Fountain, A.W., et al., Recent Advances and Remaining Challenges for the
Spectroscopic Detection of Explosive Threats. Applied spectroscopy, 2014.
68(8): p. 795-811.
134 References
106. Pamula, V.K. and R.B. Fair. Detection of dissolved TNT and DNT in soil with
a MEMS explosive particle detector. in Proceedings of SPIE. 2000.
107. Thorne, P.G. and T.F. Jenkins, A field method for quantifying ammonium
picrate and picric acid in soil. Field Analytical Chemistry & Technology,
1997. 1(3): p. 165-170.
108. Zhu, B., et al., Enhanced transformation of TNT by Arabidopsis plants
expressing an old yellow enzyme. PLoS ONE, 2012. 7(7): p. e39861.
109. Jiamjitrpanich, W., et al., Environmental factors influencing remediation of
TNT-contaminated water and soil with nanoscale zero-valent iron particles.
Journal of Environmental Science and Health, Part A, 2010. 45(3): p. 263-
274.
110. Hawari, J., et al., Microbial degradation of explosives: biotransformation
versus mineralization. Applied Microbiology and Biotechnology, 2000.
54(5): p. 605-618.
111. Tu, R., et al., Amine-Capped ZnS−Mn2+ Nanocrystals for Fluorescence
Detection of Trace TNT Explosive. Analytical Chemistry, 2008. 80(9): p.
3458-3465.
112. Yang, J., et al., Hollow silica nanospheres containing a silafluorene-fluorene
conjugated polymer for aqueous TNT and RDX detection. Chemical
Communications, 2010. 46(36): p. 6804-6806.
113. Pushkarsky, M.B., et al., High-sensitivity detection of TNT. Proceedings of
the National Academy of Sciences of the United States of America, 2006.
103(52): p. 19630-19634.
114. Sharma, S.P. and S.C. Lahiri, Absorption spectroscopic and FTIR studies on
EDA complexes between TNT (2,4,6-trinitrotoluene) with amines in DMSO
and determination of the vertical electron affinity of TNT. Spectrochimica
Acta Part A: Molecular and Biomolecular Spectroscopy, 2008. 70(1): p. 144-
153.
115. Sohn, H., et al., Detection of Nitroaromatic Explosives Based on
Photoluminescent Polymers Containing Metalloles. Journal of the American
Chemical Society, 2003. 125(13): p. 3821-3830.
116. Ma, Y., et al., White Upconversion Luminescence Nanocrystals for the
Simultaneous and Selective Detection of 2,4,6-Trinitrotoluene and 2,4,6-
References 135
Trinitrophenol. ACS Applied Materials & Interfaces, 2014. 6(10): p. 7790-
7796.
117. Enkin, N., et al., Ag Nanocluster/DNA Hybrids: Functional Modules for the
Detection of Nitroaromatic and RDX Explosives. Nano Letters, 2014. 14(8):
p. 4918-4922.
118. Ma, Y., S. Huang, and L. Wang, Multifunctional inorganic–organic hybrid
nanospheres for rapid and selective luminescence detection of TNT in mixed
nitroaromatics via magnetic separation. Talanta, 2013. 116: p. 535-540.
119. Ma, Y. and L. Wang, Upconversion luminescence nanosensor for TNT
selective and label-free quantification in the mixture of nitroaromatic
explosives. Talanta, 2014. 120: p. 100-105.
120. Gregory, K.E., et al., Quantitative comparison of trace organonitrate
explosives detection by GC-MS and GC-ECD2 methods with emphasis on
sensitivity. Journal of Chromatographic Science, 2011. 49(1): p. 1-7.
121. Halasz, A., et al., Detection of explosives and their degradation products in
soil environments. Journal of Chromatography A, 2002. 963(1–2): p. 411-
418.
122. Oxley, J.C., et al., Detection of Explosives in Hair Using Ion Mobility
Spectrometry. Journal of Forensic Sciences, 2008. 53(3): p. 690-693.
123. Ho, M.Y., N. D’Souza, and P. Migliorato, Electrochemical Aptamer-Based
Sandwich Assays for the Detection of Explosives. Analytical Chemistry,
2012. 84(10): p. 4245-4247.
124. Yu, Y., et al., A novel homogeneous label-free aptasensor for 2,4,6-
trinitrotoluene detection based on an assembly strategy of
electrochemiluminescent graphene oxide with gold nanoparticles and
aptamer. Biosensors and Bioelectronics, 2013. 43: p. 137-142.
125. Ehrentreich-Förster, E., et al., Biosensor-based on-site explosives detection
using aptamers as recognition elements. Analytical and Bioanalytical
Chemistry, 2008. 391(5): p. 1793-1800.
126. Wang, J. and M. Pumera, Microchip flow-injection analysis of trace 2,4,6-
trinitrotoluene (TNT) using mercury-amalgam electrochemical detector.
Talanta, 2006. 69(4): p. 984-987.
136 References
127. O'Mahony, A.M., et al., Orthogonal Detection of Nitroaromatic Explosives
via Direct Voltammetry Coupled to Enzyme-Mediated Biocatalysis.
Electroanalysis, 2012. 24(9): p. 1811-1816.
128. Pon Saravanan, N., et al., Voltammetric determination of nitroaromatic and
nitramine explosives contamination in soil. Talanta, 2006. 69(3): p. 656-662.
129. Wang, J. and S. Thongngamdee, On-line electrochemical monitoring of
(TNT) 2,4,6-trinitrotoluene in natural waters. Analytica Chimica Acta, 2003.
485(2): p. 139-144.
130. Chen, J.-C., et al., Disposable electrochemical sensor for determination of
nitroaromatic compounds by a single-run approach. Analytical Chemistry,
2006. 78(11): p. 3752-3757.
131. Cizek, K., et al., Integrated explosive preconcentrator and electrochemical
detection system for 2,4,6-trinitrotoluene (TNT) vapor. Analytica Chimica
Acta, 2010. 661(1): p. 117-121.
132. Riskin, M., et al., Imprinting of Molecular Recognition Sites through
Electropolymerization of Functionalized Au Nanoparticles: Development of
an Electrochemical TNT Sensor Based on π-Donor−Acceptor Interactions.
Journal of the American Chemical Society, 2008. 130(30): p. 9726-9733.
133. Gao, D., et al., A Surface Functional Monomer-Directing Strategy for Highly
Dense Imprinting of TNT at Surface of Silica Nanoparticles. Journal of the
American Chemical Society, 2007. 129(25): p. 7859-7866.
134. Wilson, R., C. Clavering, and A. Hutchinson, Electrochemiluminescence
enzyme immunoassay for TNT. Analyst, 2003. 128(5): p. 480-485.
135. Green, T.M., P.T. Charles, and G.P. Anderson, Detection of 2,4,6-
trinitrotoluene in seawater using a reversed-displacement immunosensor.
Analytical Biochemistry, 2002. 310(1): p. 36-41.
136. Oh, J.-W., et al., Biomimetic virus-based colourimetric sensors. Nature
Communications, 2014. 5.
137. Cennamo, N., et al., Sensors based on surface plasmon resonance in a plastic
optical fiber for the detection of trinitrotoluene. Sensors and Actuators B:
Chemical, 2013. 188: p. 221-226.
138. Riskin, M., et al., Ultrasensitive Surface Plasmon Resonance Detection of
Trinitrotoluene by a Bis-aniline-Cross-Linked Au Nanoparticles Composite.
Journal of the American Chemical Society, 2009. 131(21): p. 7368-7378.
References 137
139. Yatabe, R., T. Onodera, and K. Toko, Fabrication of an SPR Sensor Surface
with Antifouling Properties for Highly Sensitive Detection of 2,4,6-
Trinitrotoluene Using Surface-Initiated Atom Transfer Polymerization.
Sensors 2013. 13(7): p. 9294-9304.
140. Tu, N. and L. Wang, Surface plasmon resonance enhanced upconversion
luminescence in aqueous media for TNT selective detection. Chemical
Communications, 2013. 49(56): p. 6319-6321.
141. Larsson, A., et al., A novel biochip technology for detection of explosives -
TNT: Synthesis, characterisation and application. Sensors and Actuators B:
Chemical, 2006. 113(2): p. 730-748.
142. Shankaran, D.R., et al., Surface plasmon resonance immunosensor for highly
sensitive detection of 2,4,6-trinitrotoluene. Biosensors and Bioelectronics,
2005. 20(9): p. 1750-1756.
143. Strong, A.A., et al. Detection of trinitrotoluene (TNT) extracted from soil
using a surface plasmon resonance (SPR)-based sensor platform. in
Proceedings of SPIE. 1999.
144. O'Mahony, A.M. and J. Wang, Nanomaterial-based electrochemical
detection of explosives: a review of recent developments. Analytical Methods,
2013. 5(17): p. 4296-4309.
145. Tang, L., et al., Uniform and rich-wrinkled electrophoretic deposited
graphene film: a robust electrochemical platform for TNT sensing. Chemical
Communications, 2010. 46(32): p. 5882-5884.
146. Wang, J., et al., Sensitive electrochemical immunoassay for 2,4,6-
trinitrotoluene based on functionalized silica nanoparticle labels. Analytica
Chimica Acta, 2008. 610(1): p. 112-118.
147. Zhang, H.-X., et al., Electrochemical Sensor for Detecting Ultratrace
Nitroaromatic Compounds Using Mesoporous SiO2-Modified Electrode.
Analytical Chemistry, 2006. 78(6): p. 1967-1971.
148. Zang, J., et al., Electrochemical detection of ultratrace nitroaromatic
explosives using ordered mesoporous carbon. Analytica Chimica Acta, 2011.
683(2): p. 187-191.
149. Xia, X.-D. and H.-W. Huang, Using unmodified Au nanoparticles as
colorimetric probes for TNT based on their competitive reactions with
melamine. Chinese Chemical Letters, 2014. 25(9): p. 1271-1274.
138 References
150. Liang, X., et al., Colorimetric detection of melamine in complex matrices
based on cysteamine-modified gold nanoparticles. Analyst, 2011. 136(1): p.
179-183.
151. Lodha, A., et al., A smart and rapid colorimetric method for the detection of
codeine sulphate, using unmodified gold nanoprobe. RSC Advances, 2014.
4(92): p. 50443-50448.
152. Li, L., et al., Visual detection of melamine in raw milk using gold
nanoparticles as colorimetric probe. Food Chemistry, 2010. 122(3): p. 895-
900.
153. Lodha, A., et al., Melamine modified gold nanoprobe for "on-spot"
colorimetric recognition of clonazepam from biological specimens. Analyst,
2013. 138(18): p. 5411-5416.
154. Lisha, K.P., Anshup, and T. Pradeep, Enhanced visual detection of pesticides
using gold nanoparticles. Journal of Environmental Science and Health, Part
B, 2009. 44(7): p. 697-705.
155. Zhang, Z., et al., On-Site Visual Detection of Hydrogen Sulfide in Air Based
on Enhancing the Stability of Gold Nanoparticles. ACS Applied Materials &
Interfaces, 2014. 6(9): p. 6300-6307.
156. He, S., et al., Utilization of unmodified gold nanoparticles in colorimetric
detection. Science China Physics, Mechanics and Astronomy, 2011. 54(10):
p. 1757-1765.
157. Wei, W., et al., Fluorescent organic-inorganic hybrid polyphosphazene
microspheres for the trace detection of nitroaromatic explosives. RSC
Advances, 2012. 2(9): p. 3765-3771.
158. Gao, D., et al., Resonance Energy Transfer-Amplifying Fluorescence
Quenching at the Surface of Silica Nanoparticles toward Ultrasensitive
Detection of TNT. Analytical Chemistry, 2008. 80(22): p. 8545-8553.
159. Fang, Q., et al., Inverted Opal Fluorescent Film Chemosensor for the
Detection of Explosive Nitroaromatic Vapors through Fluorescence
Resonance Energy Transfer. Chemistry – A European Journal, 2009. 15(43):
p. 11507-11514.
160. Goldman, E.R., et al., A Hybrid Quantum Dot−Antibody Fragment
Fluorescence Resonance Energy Transfer-Based TNT Sensor. Journal of the
American Chemical Society, 2005. 127(18): p. 6744-6751.
References 139
161. Kartha, K.K., et al., Attogram Sensing of Trinitrotoluene with a Self-
Assembled Molecular Gelator. Journal of the American Chemical Society,
2012. 134(10): p. 4834-4841.
162. Zyryanov, G.V., et al., Chemosensors for detection of nitroaromatic
compounds (explosives). Russian Chemical Reviews, 2014. 83(9): p. 783.
163. Senthamizhan, A., A. Celebioglu, and T. Uyar, Ultrafast on-site selective
visual detection of TNT at sub-ppt level using fluorescent gold cluster
incorporated single nanofiber. Chemical Communications, 2014.
164. Morrissey, M.J., et al., Spectroscopy, Manipulation and Trapping of Neutral
Atoms, Molecules, and Other Particles Using Optical Nanofibers: A Review.
Sensors, 2013. 13(8): p. 10449-10481.
165. Ma, Y., et al., Highly Selective and Sensitive Fluorescent Paper Sensor for
Nitroaromatic Explosive Detection. Analytical Chemistry, 2012. 84(19): p.
8415-8421.
166. Almaviva, S., et al. Trace detection of explosives and their precursors by
surface enhanced Raman spectroscopy. in SPIE Proceedings. 2012.
167. Liu, X.J., et al., Ordered gold nanoparticle arrays as surface-enhanced
Raman spectroscopy substrates for label-free detection of nitroexplosives.
Talanta, 2011. 83(3): p. 1023-1029.
168. Graham, D., et al., Control of enhanced Raman scattering using a DNA-
based assembly process of dye-coded nanoparticles. Nature Nanotechnology,
2008. 3(9): p. 548-551.
169. Brus, L., Noble metal nanocrystals: plasmon electron transfer
photochemistry and single-molecule Raman spectroscopy. Accounts of
Chemical Research, 2008. 41(12): p. 1742-1749.
170. Enustun, B.V. and J. Turkevich, Coagulation of colloidal gold. Journal of the
American Chemical Society, 1963. 85(21): p. 3317-3328.
171. Turkevich, J., Colloidal gold. Part II. Gold Bulletin, 1985. 18(4): p. 125-131.
172. Artamkina, G.A., M.P. Egorov, and I.P. Beletskaya, Some aspects of anionic
.sigma.-complexes. Chemical Reviews, 1982. 82(4): p. 427-459.
173. Han, Y., et al., Effect of oxidation on surface-enhanced Raman scattering
activity of silver nanoparticles: A quantitative correlation. Analytical
Chemistry, 2011. 83(15): p. 5873-5880.
140 References
174. Gupta, N. and R. Dahmani, AOTF Raman spectrometer for remote detection
of explosives. Spectrochimica Acta Part A: Molecular and Biomolecular
Spectroscopy, 2000. 56(8): p. 1453-1456.
175. Sang Yun, L., et al., Solvent Effect on Formation of Cysteamine Self-
Assembled Monolayers on Au(111). Japanese Journal of Applied Physics,
2003. 42(1R): p. 236-241.
176. Kudelski, A., Characterization of thiolate-based mono- and bilayers by
vibrational spectroscopy: A review. Vibrational Spectroscopy, 2005. 39(2): p.
200-213.
177. Moskovits, M., Surface selection rules. Journal of Chemical Physics, 1982.
77(9): p. 4408-4416.
178. Creighton, J.A., Surface raman electromagnetic enhancement factors for
molecules at the surface of small isolated metal spheres: The determination
of adsorbate orientation from sers relative intensities. Surface Science, 1983.
124(1): p. 209-219.
179. Chowdhury, J., et al., Concentration-dependent orientational changes of 2-
amino-2-thiazoline molecule adsorbed on silver nanocolloidal surface
investigated by SERS and DFT. The Journal of Physical Chemistry C, 2007.
112(1): p. 227-239.
180. Freeman, R., et al., Functionalized CdSe/ZnS QDs for the Detection of
Nitroaromatic or RDX Explosives. Advanced Materials, 2012. 24(48): p.
6416-6421.
181. Bailey, H.C., et al., Toxicity of TNT Wastewaters to Aquatic Organisms.
Volume 3. Chronic Toxicity of LAP Wastewater and 2, 4, 6-Trinitrotoluene,
1985, DTIC Document.
182. Wirde, M., U. Gelius, and L. Nyholm, Self-Assembled Monolayers of
Cystamine and Cysteamine on Gold Studied by XPS and Voltammetry.
Langmuir, 1999. 15(19): p. 6370-6378.
183. Luo, S.-C., et al., Nanofabricated SERS-active substrates for single-molecule
to virus detection in vitro: A review. Biosensors and Bioelectronics, 2014. 61:
p. 232-240.
184. Han, X.X., et al., Magnetic Titanium Dioxide Nanocomposites for Surface-
Enhanced Resonance Raman Spectroscopic Determination and Degradation
References 141
of Toxic Anilines and Phenols. Angewandte Chemie International Edition,
2014. 53(9): p. 2481-2484.
185. Sivanesan, A., et al., Nanostructured silver-gold bimetallic SERS substrates
for selective identification of bacteria in human blood. Analyst, 2014. 139(5):
p. 1037-1043.
186. Kalaivani, G., et al., Regenerative silver nanoparticles for SERRS
investigation of metmyoglobin with conserved heme pocket. RSC Advances,
2013. 3(19): p. 6839-6846.
187. Kalaivani, G., et al., Plasmon-Tuned Silver Colloids for SERRS Analysis of
Methemoglobin with Preserved Nativity. Langmuir, 2012. 28(40): p. 14357-
14363.
188. Hughes, J., et al., Ultra sensitive label free surface enhanced Raman
spectroscopy method for the detection of biomolecules. Talanta, 2014. 130: p.
20-25.
189. Haynes, C.L., A.D. McFarland, and R.P. Van Duyne, Surface-enhanced:
Raman spectroscopy. Analytical Chemistry, 2005. 77(17): p. 338 A-346 A.
190. Kneipp, K., et al., Surface-enhanced Raman scattering and biophysics.
Journal of Physics Condensed Matter, 2002. 14(18): p. R597-R624.
191. Yamamoto, Y.S., Y. Ozaki, and T. Itoh, Recent progress and frontiers in the
electromagnetic mechanism of surface-enhanced Raman scattering. Journal
of Photochemistry and Photobiology C: Photochemistry Reviews, 2014. 21:
p. 81-104.
192. Anker, J.N., et al., Biosensing with plasmonic nanosensors. Nature Materials,
2008. 7(6): p. 442-453.
193. Stiles, P.L., et al., Surface-enhanced Raman spectroscopy, in Annual Review
of Analytical Chemistry2008. p. 601-626.
194. Yamamoto, Y.S., et al., Direct conversion of silver complexes to nanoscale
hexagonal columns on a copper alloy for plasmonic applications. Physical
Chemistry Chemical Physics, 2013. 15(35): p. 14611-14615.
195. Jiang, J., et al., Novel One-pot Fabrication of Lab-on-a-Bubble@Ag
Substrate without Coupling-agent for Surface Enhanced Raman Scattering.
Sci. Rep., 2014. 4.
196. Clark, A.W. and J.M. Cooper, Nanogap ring antennae as plasmonically
coupled serrs substrates. Small, 2011. 7(1): p. 119-125.
142 References
197. Yang, M., et al., SERS-active gold lace nanoshells with built-in hotspots.
Nano Letters, 2010. 10(10): p. 4013-4019.
198. Chu, H., et al., Decoration of Gold Nanoparticles on Surface-Grown Single-
Walled Carbon Nanotubes for Detection of Every Nanotube by Surface-
Enhanced Raman Spectroscopy. Journal of the American Chemical Society,
2009. 131(40): p. 14310-14316.
199. Sivanesan, A., et al., Electrochemical pathway for the quantification of SERS
enhancement factor. Electrochemistry Communications, 2014. 49(0): p. 103-
106.
200. Jamil, A.K.M., et al., Rapid detection of TNT in aqueous media by selective
label free surface enhanced Raman spectroscopy. Talanta, 2015. 134: p. 732-
738.
201. Sivanesan, A. and S. Abraham John, Amino group position dependent
orientation of self-assembled monomolecular films of
tetraaminophthalocyanatocobalt(II) on Au surfaces. Langmuir, 2008. 24(5):
p. 2186-2190.
202. Choi, S., M. Ahn, and J. Kim, Highly reproducible surface-enhanced Raman
scattering-active Au nanostructures prepared by simple electrodeposition:
Origin of surface-enhanced Raman scattering activity and applications as
electrochemical substrates. Analytica Chimica Acta, 2013. 779: p. 1-7.
203. Lv, Z.Y., et al., Shaped-controlled electrosynthesis of gold nanodendrites for
highly selective and sensitive SERS detection of formaldehyde. Sensors and
Actuators, B: Chemical, 2014. 201: p. 92-99.
204. Félidj, N., et al., Enhanced substrate-induced coupling in two-dimensional
gold nanoparticle arrays. Physical Review B, 2002. 66(24): p. 245407.
205. Shin, K. and H. Chung, Wide area coverage Raman spectroscopy for reliable
quantitative analysis and its applications. Analyst, 2013. 138(12): p. 3335-
3346.
206. Shin, K., et al., Au nanoparticle-encapsulated hydrogel substrates for robust
and reproducible SERS measurement. Analyst, 2013. 138(3): p. 932-938.
207. He, X., et al., ZnO-Ag hybrids for ultrasensitive detection of trinitrotoluene
by surface-enhanced Raman spectroscopy. Physical Chemistry Chemical
Physics, 2014. 16(28): p. 14706-14712.
References 143
208. Sil, S., et al., Density Functional Theoretical Modeling, Electrostatic Surface
Potential and Surface Enhanced Raman Spectroscopic Studies on
Biosynthesized Silver Nanoparticles: Observation of 400 pM Sensitivity to
Explosives. The Journal of Physical Chemistry A 2014. 118(16): p. 2904-
2914.
209. Sivanesan, A., et al., Tailored silica coated Ag nanoparticles for non-invasive
surface enhanced Raman spectroscopy of biomolecular targets. RSC
Advances, 2012. 2(3): p. 805-808.
210. Botta, R., et al., Silver nanoclusters films for single molecule detection using
Surface Enhanced Raman Scattering (SERS). Materials chemistry and
Physics, 2013. 137(3): p. 699-703.
211. Charles, P.T., et al., Enhanced biosensor performance for on‐ site field
analysis of explosives in water using solid‐ phase extraction membranes.
Field Analytical Chemistry & Technology, 2001. 5(6): p. 272-280.
212. Smith, K.D., et al., Detection of Smokeless Powder Residue on Pipe Bombs
by Micellar Electrokinetic Capillary. Electrophoresis. Journal of forensic
sciences, 1999. 44: p. 789-794.
213. Ayoub, K., et al., Application of advanced oxidation processes for TNT
removal: a review. Journal of hazardous materials, 2010. 178(1): p. 10-28.
214. Lima, D.R.S., et al., Impact of ammunition and military explosives on human
health and the environment. Reviews on environmental health, 2011. 26(2):
p. 101-110.
215. Van Dillewijn, P., et al., Bioremediation of 2, 4, 6-trinitrotoluene by bacterial
nitroreductase expressing transgenic aspen. Environmental science &
technology, 2008. 42(19): p. 7405-7410.
216. Bolt, H.M., et al., Genotoxicity and potential carcinogenicity of 2, 4, 6-
trinitrotoluene: structural and toxicological considerations. Reviews on
environmental health, 2006. 21(4): p. 217-228.
217. Woollen, B.H., et al., Trinitrotoluene: assessment of occupational absorption
during manufacture of explosives. British journal of industrial medicine,
1986. 43(7): p. 465-473.
218. Wang, J., Electrochemical sensing of explosives. Electroanalysis, 2007.
19(4): p. 415-423.
144 References
219. Miura, N., et al., High-performance surface plasmon resonance
immunosensors for TNT detection. Electrochemistry, 2007. 75(1): p. 13-22.
220. Gaurav, et al., SPME-HPLC: A new approach to the analysis of explosives.
Journal of Hazardous Materials, 2007. 147(3): p. 691-697.
221. Pumera, M., Analysis of explosives via microchip electrophoresis and
conventional capillary electrophoresis: A review. Electrophoresis, 2006.
27(1): p. 244-256.
222. Alvarez‐ Puebla, R.A. and L.M. Liz‐ Marzán, SERS detection of small
inorganic molecules and ions. Angewandte Chemie International Edition,
2012. 51(45): p. 11214-11223.
223. Cialla, D., et al., Surface-enhanced Raman spectroscopy (SERS): progress
and trends. Analytical and bioanalytical chemistry, 2012. 403(1): p. 27-54.
224. Sivanesan, A., et al., Complementary surface-enhanced resonance Raman
spectroscopic biodetection of mixed protein solutions by chitosan-and silica-
coated plasmon-tuned silver nanoparticles. Analytical chemistry, 2012.
84(13): p. 5759-5764.
225. Jehlička, J., et al., Raman spectra of pure biomolecules obtained using a
handheld instrument under cold high-altitude conditions. Analytical and
bioanalytical chemistry, 2010. 397(7): p. 2753-2760.
226. Kurouski, D. and R.P. Van Duyne, In Situ Detection and Identification of
Hair Dyes Using Surface-Enhanced Raman Spectroscopy (SERS). Analytical
chemistry, 2015. 87(5): p. 2901-2906.
227. Hoppmann, E.P., W.Y. Wei, and I.M. White, Highly sensitive and flexible
inkjet printed SERS sensors on paper. Methods, 2013. 63(3): p. 219-224.
228. Ang, X.F., et al. Stability of Self-Assembled Monolayers on Gold for
MEMS/NEMS Applications. in ASME 2008 International Mechanical
Engineering Congress and Exposition. American Society of Mechanical
Engineers.
229. Karsenty, M., S. Rubin, and M. Bercovici, Acceleration of surface-based
hybridization reactions using isotachophoretic focusing. Analytical
chemistry, 2014. 86(6): p. 3028-3036.
230. Jamil, A.K.M., et al., A homogeneous surface-enhanced Raman scattering
platform for ultra-trace detection of trinitrotoluene in the environment.
Analytical Methods, 2015. 7(9): p. 3863-3868.
References 145
231. Marmisolle, W.A., et al., Self-Assembled Monolayers of NH2-Terminated
Thiolates: Order, p K a, and Specific Adsorption. Langmuir, 2013. 29(17): p.
5351-5359.
232. Wang, G.-X., et al., Structure orientation of hemin self-assembly layer
determining the direct electron transfer reaction. Chemical Communications,
2015. 51(4): p. 689-692.
233. Sivanesan, A., P. Kannan, and S.A. John, Electrocatalytic oxidation of
ascorbic acid using a single layer of gold nanoparticles immobilized on 1, 6-
hexanedithiol modified gold electrode. Electrochimica Acta, 2007. 52(28): p.
8118-8124.
234. Ciobanu, M., et al., Electrochemistry and photoelectrochemistry of
photosystem I adsorbed on hydroxyl-terminated monolayers. Journal of
Electroanalytical Chemistry, 2007. 599(1): p. 72-78.
235. Kawaguchi, T., et al., Surface plasmon resonance immunosensor using Au
nanoparticle for detection of TNT. Sensors and Actuators B: Chemical, 2008.
133(2): p. 467-472.
236. Chai, F., et al., L-cysteine functionalized gold nanoparticles for the
colorimetric detection of Hg2+ induced by ultraviolet light. Nanotechnology,
2010. 21(2): p. 025501.
237. Olah, G.A. and H. Mayr, Carbanions. II. Carbon-13 nuclear magnetic
resonance study of Meisenheimer complexes and their charge distribution
pattern. The Journal of Organic Chemistry, 1976. 41(21): p. 3448-3451.
238. Pensa, E., et al., New insight into the electrochemical desorption of
alkanethiol SAMs on gold. Physical Chemistry Chemical Physics, 2012.
14(35): p. 12355-12367.
239. Lai, L.-J., et al., Surface characterization of immunosensor conjugated with
gold nanoparticles based on cyclic voltammetry and X-ray photoelectron
spectroscopy. Colloids and Surfaces B: Biointerfaces, 2009. 68(2): p. 130-
135.
240. Stine, R. and D.Y. Petrovykh, Oriented self-assembled monolayers of
bifunctional molecules on InAs. Journal of Electron Spectroscopy and Related
Phenomena, 2009. 172(1): p. 42-46.
241. Roelfs, B., et al., Adsorption of thymine on gold single-crystal electrodes. The
Journal of Physical Chemistry B, 1997. 101(5): p. 754-765.
146 References
242. Stan, D., et al., Electrochemical studies of homogeneous self-assembled
monolayers versus mixed self-assembled monolayers on gold electrode for
“label free” detection of heart fatty acid binding protein. Thin Solid Films,
2012. 526: p. 143.
243. Ro, K.S., et al., Solubility of 2, 4, 6-trinitrotoluene (TNT) in water. Journal of
Chemical & Engineering Data, 1996. 41(4): p. 758-761.