Talanta 67 (2005) 377–382

Extraction of capsaicins in aerosol defense sprays from fabrics

Oliver Spicer Jr. a , José R. Almirall b,∗
a Miami-Dade Police Crime Laboratory, Miami, FL 33172, USA
b Department of Chemistry and Biochemistry, International Forensic Research Institute,
Florida International University, University Park, Miami, FL 33199, USA

Abstract

The use of aerosol defense sprays has increased as a means of self-defense and as a weapon in the commission of a crime. The residue of
these sprays is often left behind as physical evidence on a victim’s clothing or personal belongings. As the popularity of self-defense weaponry
increases, so does the likelihood that it will be encountered in forensic casework. The extraction, recovery from fabrics, and identification
of residue from defense sprays is described. The commonly used extraction method of liquid–liquid extraction is compared to solid phase
microextraction (SPME) to recover capsaicin and dihydrocapsaicin from cotton swabs. The use of SPME resulted in lower limits of detection
and greater recovery efficiency when compared to solvent extraction. SPME also provided more consistent recovery and less variability when
compared to solvent extraction. The effect of use of various types of evidence packages on the preservation of this type of evidence is also
reported. The collection and analysis of hand swabs after normal discharge of pepper spray canisters was studied indicating the low persistence
of these compounds on the hands of the person conducting the spraying. Finally, the results of a real case whereby solvent–solvent extraction
did not provide the necessary sensitivity for extracting the capsaicin compounds on the garments of a victim of an alleged spraying and the
SPME extraction provided the recovery and identification of the compounds is also presented.
© 2005 Published by Elsevier B.V.

Keywords: Capsaicin; Dihydrocapsaicin; Pepper sprays; Aerosol defense sprays; SPME

1. Introduction charge at airports, causing the closure of entire concourses, to

claims of abuse by law enforcement used in dispersing large
Aerosol defense sprays are compounds that cause tem- groups of protesters. Nationally televised and reported cases
porary incapacitation by producing sensory irritation. As a of excessive use of defense sprays have led to greater atten-
result, extreme discomfort or pain is associated with the areas tion by the public. The World Trade Organization protests in
affected. The nose, eyes, and respiratory tract are the pri- Seattle, the apprehension of immigrants at sea in Miami, dis-
mary organs affected. The three basic components, which charge of pepper spray in a Chicago nightclub, and protesting
make up aerosol defense sprays, are: the active ingredient the destruction of Northern California’s old growth forests
(irritant), carrier, and propellant. The carrier acts as a vehi- has increased the public’s awareness concerning the use of
cle in which the irritant is suspended or dissolved. Alcohol, defense sprays. Pepper spray cases have also found their way
organic hydrocarbons, and methylene chloride are examples into criminal and civil courtrooms.
of carriers used in sprays [1,2]. Propellants are used to expel
the irritant from the canister. Commonly used propellants, 1.1. Oleoresin capsicum (OC)
include butane, propane, and compressed gases (e.g. carbon
dioxide or nitrogen) [2]. The latest addition to the list of active agents in defense
Aerosol defense sprays have garnered significant media sprays is oleoresin capsicum, classified as an inflammatory
attention recently. Their use has ranged from accidental dis- agent and the primary analyte of interest in this paper. Expo-
sure to oleoresin capsicum (OC) produces inflammation and
∗ Corresponding author. swelling of the mucous membranes associated with the eyes,
E-mail address: almirall@fiu.edu (J.R. Almirall). nose, and throat. OC’s inflammatory properties reportedly

0039-9140/$ – see front matter © 2005 Published by Elsevier B.V.

doi:10.1016/j.talanta.2005.05.031
378 O. Spicer Jr., J.R. Almirall / Talanta 67 (2005) 377–382

render the agent more effective than chloroacetophenone pounds from fabrics is required in order to identify capsaicin
(CN) and o-chlorobenzylidene malononitrile (CS) on violent, compounds in a pepper spray case. Previous efforts to extract
impaired, and mentally ill individuals. OC is a reddish- capsaicin compounds from cotton, wool, nylon, and other
brown oily liquid derived from the plants of the genus fabrics have all involved liquid–liquid solvent extraction to
capsicum, commonly referred to as hot peppers or chilies recover capsaicin. Lewis et al. compared four solvents for
[3]. OC contains a group of compounds called capsaicinoids, the recovery of 2-chlorobenzylidenemalononitrile (CS) and
which are responsible for the pungency associated with capsaicin from cotton fabric followed by GC–MS analysis.
cayenne and other varieties of peppers. Capsaicinoids are the [7]. These authors concluded that ethylacetate was the most
pharmacologically active and pain producing components efficient solvent, although all solvents investigated resulted in
of the hot pepper [4]. The active ingredient believed to be very low recovery rates when analyzed. A similar study per-
responsible for the irritative properties of OC is capsaicin. formed by Pepler, used a solution of ethylacetate/heneicosane
The second most common capsaicinoid is dihydrocapsaicin. to extract spiked samples from cotton, denim, fake leather,
Five naturally occurring analogues of capsaicin have been chenille, wool and courduroy [8]. Capsaicin was success-
reportedly isolated from pepper plants [4] (see Fig. 1 fully separated and identified by GC–MS analysis following
for structures). OC contains over 100 distinct volatile extraction. This study noted that the fake leather material
components in addition to capsaicin [5]. The exact chemical interfered with detecting trace levels. The study also included
composition of OC varies with the type of pepper, its age, results on the persistence of capsaicin residue on fabric when
and parts of the plant from which the extract is obtained. samples were properly stored, although no recovery efficien-
The basic chemical structure of capsaicin and its ana- cies were reported to assess the effectiveness of the solvent
logues is a 4-hydroxy-3-methoxybenzylamide, connected to solution. Finally, Reilly et al. spiked cotton, wool, blended,
an acyl chain containing 10–11 carbon atoms. Pure capsaicin and nylon fabric samples with pepper spray (containing
is insoluble in water, but soluble in oils and some solvents. ∼0.5 mg capsaicinoids) and extracted each with n-butyl chlo-
Capsaicin forms crystals, has a melting point of 65 ◦ C and ride followed by liquid chromatography–mass spectrometry
a boiling point at 210–220 ◦ C. Capsaicin has been used in (LC–MS) analysis [6]. This study demonstrated that 85% of
neurological research to stimulate sensory nerves and also to the original concentrations of capsaicin was detectable by
treat bladder inflammation. It is also found in topical oint- LC/MS for up to 6 months after storage. The main purpose
ments used for arthritis and neuralgia. Capsaicin exerts its of this study is to develop a sensitive method for the extrac-
effect on the sensory nerves by interacting with the vanil- tion of capsaicin compounds from fabrics and compare the
loid receptor, promoting the release of substance P, as well utility and sensitivity of the method to previously methods of
as other cytokines [6]. The release of these cytokines from recovery.
the peripheral sensory neurons causes a sensation of intense Solid phase microextraction (SPME) has proven to be
burning and pain. an important sample preparation technique for the analysis
The extraction and identification of these compounds has of forensic specimens due to the many advantages that the
become increasingly important in forensic casework. Inad- technique offers when it is applied to these types of sam-
equate extraction procedures may lead to a conclusion of a ples [9]. SPME allows for multiple sampling, preservation
false negative determination. A procedure that is capable of of the sample, minimizes the risk of sample contamination
extracting and detecting very small quantities of these com- due to the simplicity of the technique, is often faster than

Fig. 1. Chemical structures of the capsaicinoids and nonivamide (IS).

 O. Spicer Jr., J.R. Almirall / Talanta 67 (2005) 377–382 379

traditional techniques and can be readily automated. Also, 2.1. Standards preparation
the lower detection limits generally afforded by SPME allow
for confirmation of positive samples that previously went Capsaicin and dihydrocapsaicin standards were spiked
undetected. An additional benefit of SPME is the elimina- onto cotton swabs by adding 100 ␮L of a methanol solu-
tion of solvents which can save forensic science laboratories tion containing 0.1 ␮g of capsaicin and dihydrocapsaicin. For
money and reduce or eliminate the risk of analysts being each method evaluated, cotton swabs spiked with standards
exposed to toxic substances. Forensic applications of SPME were used. All samples were dried overnight in a fume hood
have included the analysis of ignitable liquid residues, often at room temperature.
referred to as accelerants [10–18], trace residues of explo-
sives [19–25], drugs and poisons from biological specimens 2.2. Liquid solvent extraction conditions
[26–40], and other forensic applications.
Methanol, ethylacetate, chloroform, and methylene chlo-
ride were each used as extraction solvents in order to deter-
2. Materials and methods mine the most effective solvent for the extraction of capsaicin
and dihydrocapsaicin from cotton swabs. Twelve previously
Capsaicin (8-methyl-n-vanillyl-6-nonenamide), dihydro- spiked cotton swabs that were prepared in triplicate for each
capsaicin (8-methyl-n-vanillylnonanamide), and nonivamide solvent used contained 10 ␮g of capsaicin and dihydrocap-
(n-vanillylnonamide) were purchased from Sigma–Aldrich saicin. The swabs were placed individually into separate
Chemicals (St. Louis, Missouri). All solvents used were of 13 mm × 100 mm test tubes. Two milliliter of each solvent
analytical grade and purchased from Fisher Scientific (Fair- were added to the test tube and sonicated for 20 min. The sol-
lawn, NJ). Pepper spray canisters were purchased locally vent was decanted and, with the aid of a pipette tip, the cotton
from various police supply vendors. Spices and topical prod- swabs were pushed through the tip to express any remain-
ucts were purchased from local grocers and pharmacies. ing solvent from the swabs. To each test tube, 100 ␮L of
The solid phase microextraction (SPME) holder and 250 ng/␮L of nonivamide was added as an internal standard
fibers (100 ␮m polydimethylsiloxane (PDMS), 65 ␮m (IS). The extracts were then reduced to dryness under a stream
polydimethylsiloxane/divinylbenzene (PDMS/DVB), 85 ␮m of nitrogen, and reconstituted with 100 ␮L of methanol. The
carboxenTM /polydi-methylsiloxane (CAR/PDMS), 70 ␮m reconstituted extract was then transferred to an autosample
carbowax® /divinylbenzene (CW/DVB), and 50/30 ␮m vial containing an insert for analysis. The final concentra-
divinylbenzene/carboxen/PDMS (DVB/CAR/PDMS)) were tion of the internal standard in the reconstituted extract is
purchased from Supelco (Bellefonte, PA). Autosampler 250 ng/␮L.
vials, rubber septa, glass vials, and 13 mm × 100 mm dispos-
able glass tubes were all purchased from Fisher Scientific 2.3. SPME extraction and fiber selection
(Pittsburgh, PA).
Analytical standards were prepared by weighing the All the SPME fibers were conditioned according to man-
appropriate quantity of capsaicin and dihydrocapsaicin using ufacture’s conditioning recommendations prior to use. Fiber
a Mettler AE 160 analytical balance to prepare a 1 mg/mL selection was experimentally determined by comparing the
solution. Stock solutions of 1, 10, 100, 500, and 1000 ng/␮L results of the extraction of a known concentration of sam-
were prepared by serial dilution in methanol for each analyt- ples in duplicate for each fiber type. The cotton swabs that
ical technique used. were previously spiked with capsaicin and dihydrocapsaicin
Analysis of capsaicin and dihydrocapsaicin was per- and dried over night were used to determine the initial recov-
formed using an Agilent (Hewlett-Packard) 5890 equipped ery of the analytes. The ends of the applicator sticks were
with a 5970 mass selective detector (Agilent Technologies, cut off and each swab was placed individually in a 10 mL
Palo Alto, CA). Separation of the analytes and internal stan- vial. Three milliliter of a 10% aqueous solution of methanol
dard was achieved using a J&W (Agilent Technologies, Palo containing 25 ␮g/mL (ppm) nonivamide as an internal stan-
Alto, CA) HP-5MS, 30 m × 0.25 mm i.d., 0.25 ␮m column. dard was added to the vial and sealed with a crimp cap. A
The gas chromatograph was equipped with an auto sam- ring stand was used to hold a SPME needle vertically above
pler with the injection volume set to 1 ␮L. The mass spec- a water bath under sonication. The rubber septum on the
trometer was operated in the full scan mode from 40 to 10 mL vial was pre-pierced to prevent bending of the SPME
400 amu. The chromatographic conditions included an initial needle. The vial containing the swab and internal standard
oven temperature of 150 ◦ C with no hold and a temperature mixture was then placed inside a beaker and immersed in the
ramp of 20 ◦ C/min to 200◦ , followed by a temperature ramp water bath. The SPME holder was positioned above the vial
of 10.0 ◦ C/min, until the final temperature of 280 ◦ C was and the needle was inserted into the pre-pierced hole. The
reached and held for 5 min. The injection port and transfer SPME fiber was then directly exposed to the aqueous liq-
line temperatures were set to 250 and 280 ◦ C, respectively. uid under sonication for 20 min. The fiber was then retracted
The gas flow rate was set to 1 mL/min and a splitless injection and the SPME needle was inserted into the injection port of
was used. the GC.
380 O. Spicer Jr., J.R. Almirall / Talanta 67 (2005) 377–382

2.4. Preservation studies saicin, dihydrocapsaicin, and the internal standard are clearly
separated by GC/MS analysis. The detection limits for cap-
Long-term persistence and preservation studies were con- saicin and dihydrocapsaicin in the split-less mode using a
ducted using various types of evidence packaging in order 1 ␮L injection volume were 7.6 and 7.0 ng, respectively.
to determine the best packaging for storing evidence with The extraction efficiencies of the organic solvents methanol,
capsaicin compounds. Duplicate sets of swabs were spiked ethylacetate, methylene chloride, and chloroform, were com-
for each evidence package evaluated. Metal cans, heat-sealed pared. Methanol provided the best recovery (50–60%) for
plastic bags, and brown bags were used as preservation pack- capsaicin followed by ethylacetate (44%), chloroform (35%),
aging. Analyses were conducted at the following intervals and methylene chloride (30%). Recovery of dihydrocapsaicin
after spiking with the analyte compounds: 12 h, 2, 3, and 4 was considerably less than capsaicin for all solvents with
months. Spiked samples were dried overnight, labeled and methanol providing the best recovery (36%), followed by
stored at room temperature. ethyl acetate (35%), chloroform (32%), and methylene chlo-
ride (28%). These recoveries compare favorably with previ-
ous reports of solvent extractions where the authors reported
2.5. Hand swab studies “a very low recovery rate” for similar solvents [7]. Cali-
bration curves for the solvent extractions of capsaicin and
Volunteers were asked to discharge pepper spray canisters dihydrocapsaicin were linear from 10 to 500 ␮g/mL (ppm),
at paper silhouettes in an open field. Various types of pepper with correlation coefficients of 0.998 and 0.996, respectively.
sprays were used. Each volunteer was instructed to discharge
the canisters according to the manufacturers directions sev-
eral times in 1 s bursts. Hand swabs were collected prior to 3.2. SPME extraction
discharge, within 15 min after discharge, and at 30 min inter-
vals for up to 1.5 h. Sterile cotton applicator swabs saturated The comparison of the results for the extraction efficiency
with methanol were used to swab the left and right palms and for the different SPME fibers is shown in Fig. 3. PDMS/DVB
fingertips of each volunteer. and DVB/CAR/PDMS resulted in nearly identical recov-
ery for both capsaicin and dihydrocapsaicin. PDMS/DVB
was selected over DVB/CAR/PDMS due to the fact that the
3. Results and discussion PDMS/DVB produces less background and the fiber can be
conditioned faster than the DVB/CAR/PDMS fiber. The cal-
3.1. Solvent extraction ibration curves for capsaicin and dihydrocapsaicin using the
PDMS/DVB fiber were found to be linear for the concentra-
The chromatogram corresponding to the separation using tion range of 10–50 ␮g/mL, with correlation coefficients of
solvent extraction is shown in Fig. 2. The components cap- 0.999 and 0.996, respectively. The recovery efficiency repre-

Fig. 2. Chromatogram of: (a) nonivamide, (b) capsaicin, and (c) dihydrocapsaicin from a spiked sample extracted with methanol.
 O. Spicer Jr., J.R. Almirall / Talanta 67 (2005) 377–382 381

Table 1
Persistence based on evidence packaging
Packaging Compound Month 2 (%) Month 3 (%) Month 4 (%)
Metal can Capsaicin, dihydrocapsaicin 89.4, 100 71.0, 91.4 68.7, 71.4
Plastic bag Capsaicin, dihydrocapsaicin 89.4, 97.4 90.6, 90.1 77.9, 54.5
Brown bag Capsaicin, dihydrocapsaicin 53.5, 63.8 46.8, 63.8 33.6, 41.5
Amount recovered as percentage after time stored in packaging.

sents the initial recovery from swabs spiked with capsaicin Table 2
and dried overnight in a fume hood and determined to be 77% Results of hand swab extraction followed by SPME/GC/MS analysis of
hands of volunteers that discharged aerosol products
for capsaicin, and 53% for dihydrocapsaicin using SPME.
These recoveries are considerably better than previous reports Volunteer Left hand Right hand Aerosol product
using solvent extractions [7]. #1 − − Sabre red
#2 − + Sabre red/disabler
#3 − + 911
3.3. Persistence study #4 − − MK-4
#5 − − MK-4
#6 − − 911
Table 1 summarizes the results of the persistence study
#7 − − Sabre red #2
for capsaicin and dihydrocapsaicin when different packag-
ing materials were used. The results show that the use of All but two extractions resulted in negative results.
certain evidence packaging contributed to the persistence of
capsaicin. Metal cans that are typically used in the collec- 3.5. Case study
tion of fire debris evidence from suspicious fires had nearly
identical concentrations to heat sealed plastic bags after 2 Clothing from a case involving the alleged use of pepper
months of storage. After 2 months of storage in a brown bag, spray was submitted for analysis. Simultaneous extractions
the concentration of capsaicin was nearly reduced to half. were performed using both the solvent and SPME tech-
The concentrations after 4 months of storage in metal cans niques described above. Several sections of clothing mea-
and heat sealed plastic were still generally 70% of the origi- suring 1 cm × 1 cm were cut from the same area suspected
nal concentration and above for capsaicin. Samples stored in of having residue and analyzed. The SPME/GC/MS proce-
brown bags resulted in lower concentrations throughout the dure was sensitive enough to positively identify capsaicin and
duration of storage. dihydrocapsaicin in these samples while solvent extraction
failed to recover enough of either of these compounds.
3.4. Hand swabs

Swabs were taken from the hands of volunteers 10 min 4. Conclusions

after discharge of pepper spray canisters. SPME/GC/MS
analysis was performed on the swabs within 24 h after collec- Methanol was found to recover all three compounds
tion in brown bags. The results are summarized in Table 2. of interest in this study from swabs with an efficiency of
Capsaicin or dihydrocapsaicin was detected on two out of 50–60%. SPME extraction resulted in better recovery (>70%)
seven volunteers whose hands were swabbed after the initial and identification of lower quantities of the compounds of
discharge. interest when compared to solvent extraction. The solvent
extraction results in LODs of 7.6 and 7.0 ng for capsaicin
and dihydrocapsaicin, respectively, but more sample prepa-
ration (multiple steps) is required than the SPME method and
the extract contains more background substrate. The LODs
found for the SPME are 1.08 and 0.73 ng for capsaicin and
dihydrocapsaicin, respectively. Reduced sample preparation
is required (extraction occurs in a sealed vial in a single step)
and the extraction results in less background than with solvent
while achieving good calibration and linearity.
Hand swabs can be of value in determining recent use
of pepper sprays. Based on the results from this study, it
appears that with normal use, residue from sprays is not
likely to be deposited on the fingers of the person spraying.
Fig. 3. Comparison of the results for the extraction efficiency for the different Further studies are needed to determine long-term persis-
SPME fiber types for both capsaicin and dihydrocapsaicin. tence after use. One possible area of concern is being able
382 O. Spicer Jr., J.R. Almirall / Talanta 67 (2005) 377–382

to distinguish between capsaicin found in aerosol sprays and [13] K.G. Furton, J.R. Almirall, J. Bruna, J. Forensic Sci. 41 (1996)
capsaicin found in spices and medicinal ointments. While it is 12–22.
[14] J.R. Almirall, J. Bruna, K.G. Furton, Sci. Justice 36 (1996) 283–287.
unlikely that the amount of capsaicin found in these products
[15] Steffen, J. Pawliszyn, Anal. Commun. 33 (1996) 129–131.
would elicit a positive response, the detection of solvents and [16] J.R. Almirall, K.G. Furton, in: S.A. Wercinski (Ed.), Solid Phase
propellants normally found in defense sprays would provide Microextraction: A Practical Guide, Marcel Dekker, New York, 1999,
additional evidence that a defense spray has been used. Cap- pp. 203–216 (Chapter 7).
saicin can persist on clothing for a relatively long period of [17] J.R. Almirall, J. Wang, K. Lothridge, K.G. Furton, J. Forensic Sci.
45 (2) (2000) 461–469.
time if properly packaged and stored. Cans and heat sealed
[18] Q. Ren, W. Bertsch, J. Forensic Sci. 44 (3) (1999) 504–515.
plastic bags provide the best packaging for the collection [19] J.Y. Horng, S.D. Huang, J. Chromatogr. A 678 (1994) 313–318.
and preservation of this type of evidence. SPME/GC/MS was [20] K.P. Kirkbride, G. Klass, P.E. Pigou, J. Forensic Sci. 43 (1) (1998)
used to extract and analyze small sections of clothing con- 76–81.
taining residue from the spray of pepper spray formulations. [21] S.A. Barshick, W.H. Griest, Anal. Chem. 70 (1998) 3015–3020.
[22] J.R. Almirall, L. Wu, G. Bi, M.W. Shannon, K.G. Furton, in: K.
In one case, containing very small amounts of residue, sol-
Higgins (Ed.), Proceedings of the SPIE, vol. 3576, 1999, pp. 18–23.
vent extraction and SPME were both used as the extraction [23] K.G. Furton, L. Wu, J.R. Almirall, J. Forensic Sci. 45 (4) (2000)
methods. The solvent extraction failed to recover capsaicin 845–852.
while the reported method using SPME recovered capsaicin. [24] K.G. Furton, L.J. Myers, Talanta 54 (3) (2001) 487–500.
[25] K.G. Furton, J.R. Almirall, M. Bi, J. Wang, L. Wu, J. Chromatogr.
A 885 (2000) 419–432.
[26] L. Junting, C. Peng, O. Suzuki, Forensic Sci. Int. 97 (1998) 93–100.
References [27] T. Kumazawa, K. Watanabe, K. Sato, H. Seno, A. Ishii, O. Suzuki,
J. Forensic Toxicol. 13 (3) (1995) 207–210.
[1] National Institute of Justice Technology Assessment Program, Oleo- [28] S.W. Myung, H.K. Min, S. Kim, M. Kim, J.B. Cho, T.J. Kim, J.
resin Capsicum: Pepper Spray as a Force Alternative, US Department Chromatogr. B 716 (1998) 359–365.
of Justice, Washington, DC, 1994. [29] K. Ameno, C. Fuke, S. Ameno, H. Kinoshita, I. Ijiri, Can. Soc.
[2] R.J. Lee, R.L. Yolton, D.P. Yolton, C. Schneider, M.L. Janin, J. Am. Forensic Sci. J. 29 (2) (1996) 43–48.
Optom. Assoc. 67 (1999) 548. [30] M. Yashiki, T. Kojima, T. Miyazaki, J. Forensic Toxicol. 12 (2)
[3] M. Kataoka, Y. Seto, K. Tsuge, M. Naomi, J. Forensic Sci. 47 (2002) (1994) 120–121.
44. [31] M. Yashiki, T. Kojima, T. Miyazaki, N. Nagasawa, Y. Iwasaki, K.
[4] C.A. Reilly, D.J. Crouch, G.S. Yost, J. Forensic Sci. 46 (2001) 502. Hara, Forensic Sci. Int. 76 (1995) 169–177.
[5] C.A. Reilly, D.J. Crouch, G.S. Yost, A.A. Fatha, J. Chromatogr. A [32] F. Centini, A. Masti, I.B. Comparini, Forensic Sci. Int. 83 (1996)
912 (2001) 259. 161–166.
[6] C.A. Reilly, D.J. Crouch, G.S. Yost, D.M. Andrenyak, J. Forensic [33] H. Lord, J. Pawliszyn, Anal. Chem. 69 (19) (1997) 3899–3906.
Sci. 47 (2002) 37. [34] C. Jurado, M. Gimenez, T. Soriano, M. Menedez, M. Repetto, J.
[7] K. Lewis, R.J. Lewis, J. Forensic Sci. 46 (2001) 352. Anal. Toxicol. 24 (2000) 11–15.
[8] R.S. Pepler, P. Esseiva, O. Gueniat, Sci. Justice 38 (1998) 203. [35] H. Seno, T. Kumazawa, A. Ishii, K. Watanbe, H. Hattori, O. Suzuki,
[9] K.G. Furton, J. Wang, Y.-L. Hsu, J. Walton, J.R. Almirall, J. Chro- J. Forensic Toxicol. 15 (1) (1997) 16–20.
matogr. Sci. 38 (2000) 297. [36] B. Hall, J. Brodbelt, J. Chromatogr. A 777 (1997) 275–282.
[10] J.R. Almirall, K.G. Furton, J.C. Bruna, Proceedings of the South- [37] N. Nagasawa, M. Yashiki, Y. Iwasaki, K. Hara, T. Kojima, Forensic
ern Association of Forensic Scientists—Fall 1994 Meeting, Orlando, Sci. Int. 78 (1996) 95–102.
Florida, September 7–10, 1994. [38] A. Bermejo, R. Seara, A. dos Santos Lucas, M. Tabernero, P. Fer-
[11] K.G. Furton, J.R. Almirall, J. Bruna, J. High Resolut. Chromatogr. nandez, R. Marsili, J. Anal. Toxicol. 24 (2000) 66–69.
18 (1995) 625–629. [39] P. Okeyo, S. Rentz, N. Snow, J. High Res. Chromatogr. 20 (1997)
[12] T. Kaneko, M. Nakada, Rep. Natl. Res. Inst. Police Sci.: Res. Foren- 171–173.
sic Sci. 48 (1995) 107–111. [40] Z. Penton, Can. Soc. Forensic Sci. J. 30 (1) (1997) 7–12.