SAW arrays using dendrimers and pattern recognition to detect volatile organics Page: 1 of 3
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
SAW ARRAYS USING DENDRIMERS T
AND PATTERN RECOGNITION carxo
TO DETECT VOLATILE ORGANICS pinaco
r C2 tributy
Antonio J. Ricco, Gordon C. Osbourn, examp
Richard M. Crooks,3 Maurie E. Garcia,
Robert Peez,3 John W. Bartholomew,2 VERI
Rubel F. Martinez2, Ralph Spindler,4 analysi
and Mark E. Kaiser percept
1Microsensor R&D and 2Vision Science Depts. the m
Sandia National Laboratories approp
Albuquerque, New Mexico, 87185-1425 chemic
3Dept. of Chemistry, Texas A&M University perform
College Station, Texas, 77843-3255 discrim
Michigan Molecular Institute training
Midland, Michigan, 48640 choices
5Dendritech, Inc. group
Midland, Michigan, 48642 quantit
Chemical sensor arrays eliminate the need to develop a high- model
selectivity material for every analyte. The application of out an
pattern recognition to the simultaneous responses of different respond
microsensors enables the identification and quantification of T
multiple analytes with a small array. Maximum materials analyt
diversity is the surest means to create an effective array for sensors
many analytes , but using a single material family dimen
simplifies coating development. cluster
Dendrimeric polymers, or dendrimers, are attractive for for all
such arrays because they contain chemically tailorable (5 den
endogenous and exogenous binding sites, in addition to 99.5%
tunable permeability and other physical properties . above
Dendrimers, prepared by repetitive branching, have three accuracy
anatomical features: a central core, repetitive branch units, and calibra
terminal functional groups. Dendrimers are synthesized by or all c
adding repetitive branch units to the molecule one layer, or comply
generation, at a time in iterative growth steps. forms
Here we report the successful combination of an array of analys
six dendrimer films with mass-sensitive SAW (surface
acoustic wave) sensors to correctly identify 18 organic analytes assist
over wide concentration ranges, with 99.5% accuracy. The set multip
of materials for the array is selected and the results evaluated Lockh
using Sandia's Visual-Empirical Region of Influence (VERI) of Ene
pattern recognition (PR) technique [3,4].
We evaluated eight dendrimer films and one self-assem- Refere
bled monolayer (SAM) as potential SAW array coatings . 1. A
The region of the SAW delay line between transducers is first Chi
modified by a 200 nm-thick Au film, followed by a self- 2. R.
assembled monolayer of mercaptoundecanoic acid (MUA) 21
[1,2]. Chemical diversity is imparted to the dendrimers 3. G
through bulk-phase coupling of appropriate terminal functional C.
groups with the primary-amine-terminated dendrimer surface. 4. G
Next, reaction of the dendrimers' few remaining an
unfunctionalized amines with the MUA monolayer yields 11
amide-linked dendrimer films on the SAW devices. 11
Design and fabrication of SAW devices and the 5. G
measurement system have been described elsewhere [1,5]. M
The 97-MHz ST-quartz SAW delay lines provide short-term Sp
frequency stability of ca. 0.3 Hz, yielding a detection limit of Sol
100 pg/cm2. Re
DETANT OFw "T"S DWM,!
he 18 organic analytes we examined were: cyclohexane,
"e, i-octane, kerosene, benzene, toluene, chlorobenzene,
tetrachloride, trichloroethylene, methanol, n-propanol,
lyl alcohol, acetone, methyl isobutyl ketone,
/phosphate, and water. These analytes include multiple
les from several chemical classes.
We analyzed the SAW/dendrimer array data using the
PR technique [3,4], a newly-developed multivariate
s method based on an empirical model of human cluster
ion. VERI analysis correctly treats arbitrarily complex
response distributions, and automatically determines
ultivariate decision thresholds necessary to produce
rate class volume shapes for the array responses to any
al analyte. Appropriate class volume shapes are needed
ain correct estimates of the chemical recognition
ance of sensor arrays and to achieve the most effective
tination against unknown analytes (i.e., those not in the
g set). VERI leave-one-out analysis quantitatively
res the chemical recognition performance of alternative
of sensor films, and thus finds the optimal subset of a
of films for a set of target analytes. These results enable
ative array-element selection comparisons not possible
principal-component analysis [3,4]. Inclusion of
d sensor response degradation into the VERI leave-one-
alysis provides a comparison of the robustness of array
ses for different film combinations.
hus, the ability of an array to distinguish different
s using VERI PR hinges upon choosing an array of
s, the response of each constituting an axis in multi-
sional space, such that each analyte is represented by a
of points that is geometrically distinct from the clusters
other analytes. By choosing the best set of 6 coatings
drimers and one methyl-terminated SAM), we achieved
accuracy in identifying all of the compounds listed
over the 0.25 - 50% of p.. concentration range. This
y is maintained even with the addition to the
tion data of sensitivity losses of as much as 8% for any
f the 6 sensors. The distributions of responses are quite
ex, and are not easily converted into any of the simple
needed for statistical-based pattern-recognition (PR)
We gratefully acknowledge the excellent technical
nce of Alan W. Staton of Sandia. Sandia is a
rogram laboratory operated by Sandia Corporation, a
eed Martin Company, for the United States Department
rgy under Contract DE-AC04-94AL85000.
J. Ricco, R. M. Crooks, and G. C. Osbourn, Accts.
em. Res., 31, 289 (1998).
M. Crooks and A. J. Ricco, Accts. Chem. Res., 31,
C. Osbourn, J. W. Bartholomew, A. J. Ricco, and G.
Frye, Accts. Chem. Res., 31, 297 (1998).
. C. Osbourn, J. W. Bartholomew, A. M. Bouchard,
d R. F. Martinez, www.sandia.gov/
00/1 155Web/ 1155home.htm.
. C. Osbourn, A. J. Ricco, J. W. Bartholomew, R. F.
artinez, M. E. Garcia, R. Peez, R. M. Crooks, R.
indler and M. E. Kaiser, Technical Digest of the 1998
id-State Sensor and Actuator Workshop, Transducers
search Foundation: Cleveland (1998), pp. 174-177.
Here’s what’s next.
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Ricco, A.J.; Osbourn, G.C.; Bartholomew, J.W.; Martinez, R.F.; Crooks, R.M.; Garcia, M.E. et al. SAW arrays using dendrimers and pattern recognition to detect volatile organics, article, August 1, 1998; Albuquerque, New Mexico. (https://digital.library.unt.edu/ark:/67531/metadc701959/m1/1/: accessed April 19, 2019), University of North Texas Libraries, Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.