Forensic Science International 268 (2016) 139–144

Contents lists available at ScienceDirect

Forensic Science International

journal homepage: www.elsevier.com/locate/forsciint

Technical note

A survey of extraction solvents in the forensic analysis of textile dyes

Ethan Groves *, Christopher S. Palenik, Skip Palenik
Microtrace LLC 790 Fletcher Dr., Suite 106, Elgin, IL 60123, United States

A R T I C L E I N F O A B S T R A C T

Article history: The characterization and identification of dyes in fibers can be used to provide investigative leads and
Received 15 December 2015 strengthen associations between known and questioned items of evidence. The isolation of a dye from its
Received in revised form 25 February 2016 matrix (e.g., a textile fiber) permits detailed characterization, comparison and, in some cases,
Accepted 16 March 2016
identification using methods such as thin layer chromatography in conjunction with infrared and Raman
Available online 25 March 2016
spectroscopy. A survey of dye extraction publications reveals that pyridine:water (4:3) is among the
most commonly cited extraction solvent across a range of fiber and dye chemistries. Here, the efficacy of
Keywords:
this solvent system has been evaluated for the extraction of dyes from 172 commercially prevalent North
Forensic science
American textile dyes. The evaluated population represents seven dye application classes, 18 chemical
Trace evidence
Textile fibers classes, and spans nine types of commercial textile fibers. The results of this survey indicate that 82% of
Extraction the dyestuffs studied are extractable using this solvent system. The results presented here summarize
Dyes the extraction efficacy by class and fiber type and illustrate that this solvent system is applicable to a
Pyridine:Water wider variety of classes and fibers than previously indicated in the literature. While there is no universal
solvent for fiber extraction, these results demonstrate that pyridine:water represents an excellent first
step for extracting unknown dyes from questioned fibers in forensic casework.
ß 2016 Elsevier Ireland Ltd. All rights reserved.

1. Introduction difficult to evaporate, [5_TD$IF]making it ill-suited for thin layer chromatog-

In the forensic science community, dye analysis is largely
focused around textile fibers, which are among the most
commonly encountered type of trace evidence [1]. The application
of modern analytical techniques can aid to classify and, in some
cases, identify individual dyestuffs [2–5]. A combination of matrix
effects, concentration limitations, and the presence of multiple
dyes in a single textile1[1_TD$IF] can limit the practical feasibility of in situ
dye identification. Thus, in order to characterize a dyestuff, it is
generally necessary to isolate it from the substrate, preferably by
means that do not alter the dyestuff or destroy the matrix.
It is understood, and perhaps expected, that no extraction
protocol can be inclusive of all dyestuff chemistries or analytical
techniques. For example, the solvent of choice used by the
Netherlands Forensic Institute (NFI) is dimethyl sulfoxide (DMSO)
[6]. This solvent works well for the extraction of many dyes, and the
solution can be directly utilized for HPLC analysis. Unfortunately,
DMSO, with its low volatility and relatively high boiling point, is
raphy (TLC) analyses.
This research was conducted in order to identify an efficient and
widely applicable extraction approach consistent with forensic
limitations, which include: casework-sized samples (fragments of
single fibers, which may be as small as a few millimeters in length
depending on the depth of shade) [7], unknown dye(s), and the
need to remove the solvent system from the extracted dye(s). To
approach this question, we conducted a survey of the literature to
identify the range of solvent systems that are used to extract dyes
from fibers and[3_TD$IF] develop an understanding of the fibers and dye
classes to which those solvent systems had been successfully
applied. Next, a population of commercially relevant dyes was
identified by surveying manufacturers for the dyes that are
currently commercially produced for textile applications. A
comparison of the data suggested that pyridine:water would
serve as a broadly efficacious solvent system for the goals of this
project. While this solvent system is not novel in and of itself, its
extraction efficacy was systematically evaluated over a wide range
of dyed fibers spanning various fiber types and dye classes.

2. Materials and methods

* Corresponding author. Tel.: +1 8477429909.
E-mail address: Egroves@microtracellc.com (E. Groves).
1
This casework observation is supported by our contacts within the trade 2.1. Extraction approach
industry who confirm that, when possible, inexpensive colorants are used to
generate the base color of a product, which is then adjusted with small amounts of A great number of extraction procedures for dye classification
more expensive colorants to achieve the desired end color. and identification are available in the literature. Many early

http://dx.doi.org/10.1016/j.forsciint.2016.03.043
0379-0738/ß 2016 Elsevier Ireland Ltd. All rights reserved.
140 E. Groves et al. / Forensic Science International 268 (2016) 139–144

Table 1
Literature review of proposed dye extraction solvent systems applied to fibers (listed alphabetically).

Suggested solvent systems

1 Acetic acid 19 Formic acid

2 Acetic acid and chlorobenzene 20 Formic acid and DMSO
3 Acetone and water 21 Formosul G and Developer O
4 Alcohol 22 Hydrochloric acid
5 Alcoholic ammonia 23 Methyl acetate and water
6 Ammonia 24 Methylene chloride
7 Aniline 25 Nitrobenzene
8 Caprolactam and ether and water 26 N-methyl-2-pyrrolidone (aq.)
9 Cellulase 27 Oxalic acid and pyridine and water
10 Chlorobenzene 28 Pyridine and water
11 Chloroform 29 Sodium carbonate
12 DMF 30 Sodium carbonate and sodium sulfide
13 DMF and chloroform 31 Sodium dithionite
14 DMF and formic acid 32 Sodium hydroxide
15 DMF and water 33 Sodium hydroxide and ammonium chloride
16 DMSO 34 Sodium hydroxide and sodium dithionite
17 Ethylenediamine 35 Sulfuric acid and ammonia
18 Formamid 36 Toluene

extraction protocols were developed for the textile industry. They
have since been expounded upon to incorporate the ever growing
population of synthetic dyes, and, some of these have also been
developed with considerations specific to forensic investigations.
Possibly the most commonly cited early forensic publications on
dye analysis arise from the Home Office Central Research
Establishment (HOCRE) in the late 1970’s to early 1980’s [8–
12]. While this work stemmed from earlier publications [13–17]; it
ultimately led to subsequent and broader publications [18–23]2 as
well as the development of standard methods [24]. While some
extraction methods have been developed for a specific subset of
dyes, others build on previously reported methods that span a
wider or updated set of dyestuffs. Surveying multiple, albeit closely
related, references provides insight to the progressive develop-
ment of dye extraction methods. A summary of various extraction
systems, which has been compiled from the literature, is presented
in Table 1.
Condensing the extraction schemes into a simple summary of
solvents does not reflect the complexity of these methods, many of
which consider both the concentration of the various solvents or
the serial steps that define an extraction approach (e.g., pretreat-
ment with oxalic acid followed by extraction with pyridine:water).
Despite some loss of resolution, a summary of the solvent systems
by application class and fiber type (Table 2) does provide a means
by which to assess the breadth of a particular solvent system’s
applicability. Upon examination of Table 2, it is apparent that
pyridine:water has a broad base of applicability across both fiber
compositions and dyestuff chemistries.
2.2. Selection of commercially relevant textile dyes
To evaluate the efficacy of an extraction solvent it seemed
prudent to examine dyes that are of current commercial importance
and use. Global Industry Analysts, Inc. reported in 2011 (shortly
before the commencement of this work in 2012), that just over half
the market share of dyestuff revenue belonged to three major
companies: BASF, Clariant, and Dystar [25]. A list of commercially
available dyestuffs was assembled based upon a compilation of the
North American product offerings from these manufacturers [7_TD$IF]as well
as the list of commercially prevalent dyestuffs provided in the
appendix of Industrial Dyes [26]. This list was examined for
recurring dyestuffs to produce a list of commercially common
dyestuffs. The resulting sample set is anticipated to provide a
2
Only a handful of publications are referenced. There is a wealth of literature
regarding dye analysis that is far too numerous to detail here.
reasonable distribution of common and currently available dyes
predominately marketed toward textile applications in North
America. It is important to note, that only dyestuffs which have
Colour Index (C.I.) classifications were included in this survey.
While dyes without C.I. certainly cannot be ignored, little, if any,
chemical information is generally available for such dyes.
2.3. Fiber sample selection
When considering fiber extraction efficacy, the matrix (i.e., fiber
type) in which a dye is present is nearly as important as the
chemistry and commercial prevalence of the dye. Not only does the
fiber chemistry impact the type and range of dyes which may be
present, but the composition of the fiber may have an impact on
the extraction method(s) employed. In addition to the extraction
scheme developed at the Metropolitan Police Forensic Science
Laboratory, Wiggins also compiled a list of fiber types in which a
particular class of dye was encountered (Table 3) [18]. This
provides an idea of the range of dye classes which may be
encountered on a given fiber type. During the course of sample
selection, care was taken to encompass the range of fiber[8_TD$IF]-dye class
combinations represented in this table.
A total of 195 fiber samples, spanning nine fiber classes, and
representing 172 unique and commercially produced dyes are
represented in the sample set studied here. A summary of the
samples by dye class and fiber type is provided in Table 4. The
samples all originate from shade cards,3[6_TD$IF] and the information
provided on them by the manufacturer was assumed to be correct.
The selected fiber samples include yarns and fabrics, printed and
dyed, spanning a range of dyestuff concentrations. When multiple
examples of a fiber[8_TD$IF]-dye combination were present in our collection,
often the darkest example was selected, in anticipation that a color
change as the result of extraction would be most easily visualized.
One intentional simplification of this study was the use of fibers that
contain only a single dye, which allows for the extraction of a
specific dyestuff to be evaluated. Additionally, the authors refrained
from using fabrics/yarns of mixed fiber composition.
2.4. Extraction setup
A pyridine (99.9%, Fisher Scientific) and water (distilled,
bottled) (4:3, v/v) solvent system was selected for evaluation.
3
The term ‘‘shade card’’ refers to marketing materials produced by colorant
manufacturers for the purpose of displaying their products, typically as specified
dyes, at specified colorant concentrations (depth of shade), on specific fabrics or
fiber types.
 E. Groves et al. / Forensic Science International 268 (2016) 139–144 141

This decision was based upon the literature survey which suggests

specified
that this system is the most widely applicable and practical (for

Not

19

31
31
many forensic applications).
For this work, a single tuft of yarn, approximately 1 cm in
length, was placed into the bottom of a 1/2 dram, clear glass vial
Vegetable (Glassvials.com, V1235C-TFE), along with 300 mL of the extraction
fibers solvent. The vial was sealed and placed in an oven at 100 8C for

19

19
6 30 min. While this sample volume is larger than a typical
questioned sample in forensic casework, it simplifies the evalua-
27, 28

tion of extraction efficacy. In contrast, questioned fibers in

23
casework often consist of only a single fiber fragment. In such
PP

instances, an equivalent amount of the potential known sample

should be extracted4 to determine (a) the efficacy of the extraction
16, 29

6, 31
5, 6,

solvent and (b) the minimum length of fiber needed to obtain a

1, 4
Silk

34
6

useful quantity5 of extracted colorant. In such cases, the smallest

Suggested solvent systems presented in Table 1 arranged by dye application class and fiber type. Applications of pyridine:water (system 28) are highlighted in boldface text.

possible volume of extraction solvent should be utilized (typically

16, 27, 28, 29

just enough to cover the sample, which has been pressed into the

2, 12, 16, 28
1, 4, 21, 31

bottom of a small conically shaped vial/tube).

7, 11, 34
5, 6, 15,

After the 30 min extraction period, the vials were removed from
17, 27
Wool

the oven and qualitatively evaluated using a stereomicroscope to

32

determine the effectiveness of the extraction. Qualitative rankings

of extraction are not uncommon in the literature. For example,
12, 15, 16, 28
6, 15, 16, 28

Schweppe [17] evaluated several extraction solvents with ‘‘fair,’’

28, 31, 34
28, 30, 31
1, 13, 22,
1, 21, 33

9, 32, 35

‘‘good,’’ and ‘‘little stripped’’ descriptors. Though, somewhat more

Cotton

standardized is the numerical, semi-quantitative approach de-

1, 28

scribed in the literature [8,9] whereby samples are ranked on a

1

scale from one to five: a value of one represents no extraction, three

a moderate extraction, in which the amount of dye in solution is
6, 15, 16, 28

28, 31, 34
28, 30, 31

comparable with that remaining in the fiber, and five being a

1, 13, 22,
9, 32, 35

complete extraction of the dye from the fiber. Rankings of two and
Rayon

1, 28
1, 33

four represent instances where the dye concentration favors the

15
1

fiber or solution, respectively.

Fibers with a ranking of one or two (no or poor extraction) were
2, 12, 16, 25
8, 10, 12, 16

returned to the oven for an additional 30 min in attempt to

8, 14, 28

improve an extraction. Samples ranking three or four (moderate to

fair, but not complete, extraction) were placed in a new aliquot of
PET

extraction solution to minimize the potential for saturation of the

extraction solution and returned to the oven for an additional
6, 15, 16, 28, 29

30 min extraction period. Following the extraction process

2, 14, 15, 16,

described above, the fibers were visually examined to rate the

18, 25, 28
2, 10, 28

extraction efficiency.
28, 29

28, 29
Nylon

28

28

3. Results and discussion

2, 15, 19, 26
Acrylic and

A summary of the samples evaluated and their extraction

modacrylic

15, 19, 20,

19, 26, 28

18, 19, 28

efficiencies are presented in Table 5.

26, 28

A visually complete extraction was observed in 83 (43%) of the

34

samples after the first 30 min extraction period in pyridine:water

(4:3). This number increased to 118 (61%) after the second
3, 10, 15, 16,

30 min extraction period. It is possible that additional time and/or

3, 4, 10, 24,
Fiber type

24, 26, 28

26, 28, 36

an additional aliquot of extraction solvent could further improve

Acetate

19, 26

the extraction ranking of the samples. This inference is based upon

26

the general improvement of extraction between the 30 and 60 min

evaluations. For instance, 69 samples were ranked either three or
four after the first 30 min extraction period. This number
Not specified

decreased to 38 after the second extraction. Thus, 55% of the

Mordant
Disperse

Reactive

samples originally ranked three or four improved their overall

Direct

Sulfur
Azoic
Basic
Acid

extraction ranking with the additional extraction time. Conversely,

Vat

of the 43 samples originally ranked with a one or two, only 16%

(seven samples) showed any visual improvement in extraction.
Dye application

After a total of 60 min in pyridine:water (4:3), a total of

4
The test extraction applied to the known sample should include any pre-
class
Table 2

extraction steps such as cleaning to remove adhering debris or residue.

5
The minimum amount of colorant needed depends upon the follow-up
characterization anticipated for the extracted colorant.
142 E. Groves et al. / Forensic Science International 268 (2016) 139–144

Table 3
Combination of dye classes and fiber types which were encountered in casework (from Wiggins [18][4_TD$IF]) (reproduced with permission from the publisher).

Fiber type

Acetate Acrylic/Modacrylic Nylon PET* Rayon Cotton Wool Silk PP**

Dye application class Acid     

Azoic  
Basic   
Direct  
Disperse     
Ingrain 
Mordant  
Reactive   
Sulfur 
Vat 
*
PET = polyester terephthalate.
**
PP = polypropylene.

Table 4
Distribution of the dye–fiber combinations examined. Percentage is relative to fiber type.

Fiber type

Acetate Acrylic/Modacrylic Nylon PET* Rayon Cotton Wool Silk Linen # of samples

Dye application class Acid 22% 40% 42% 9% 8% 79% 67% 54

Basic 40% 3% 5% 8% 11
Direct 9% 35% 32% 29
Disperse 56% 20% 36% 91% 24% 20% 8% 25% 63
Mordant 11% 3% 3% 2% 4% 8% 6
Reactive 11% 6% 22% 41% 100% 29
Vat 3% 5% 3
# of samples 9 15 33 21 37 41 24 12 1
*
PET = Polyester terephthalate.

Table 5
List of the studied dyes and their corresponding extraction result.

C.I. Fiber type Final extraction value C.I. Fiber type Final extraction value

C.I. Acid Black 1 Wool 5 C.I. Acid Red 151 Rayon 5

C.I. Acid Black 24 Nylon 4 C.I. Acid Red 249 Silk 5
C.I. Acid Black 52 Rayon 4 C.I. Acid Red 266 Wool 5
C.I. Acid Black 60 Nylon 5 C.I. Acid Red 337 Wool 5
C.I. Acid Black 107 Nylon 5 C.I. Acid Violet 7 Nylon 5
C.I. Acid Black 172 Nylon 5 C.I. Acid Violet 12 Acrylic 5
C.I. Acid Blue 9 Wool 5 C.I. Acid Violet 43 Modacrylic 5
C.I. Acid Blue 25 Acrylic 5 C.I. Acid Yellow 1 Wool 5
C.I. Acid Blue 40 Nylon 5 C.I. Acid Yellow 3 Rayon 5
C.I. Acid Blue 45 Nylon 5 C.I. Acid Yellow 7 Wool 5
C.I. Acid Blue 62 Silk 5 C.I. Acid Yellow 11 Modacrylic 5
C.I. Acid Blue 74 Wool 5 C.I. Acid Yellow 17 Polyester 5
C.I. Acid Blue 80 Wool 5 C.I. Acid Yellow 23 Nylon 5
C.I. Acid Blue 92 Wool 5 C.I. Acid Yellow 49 Wool 5
C.I. Acid Blue 113 Wool 4 C.I. Acid Yellow 127 Wool 2
C.I. Acid Blue 129 Silk 5 C.I. Acid Yellow 219 Wool 5
C.I. Acid Blue 158 Polyester 5 C.I. Basic Blue 9 Rayon 5
C.I. Acid Blue 298 Nylon 5 C.I. Basic Green 4 Acrylic 5
C.I. Acid Blue 324 Nylon 5 C.I. Basic Orange 21 Acrylic 3
C.I. Acid Green 16 Wool 5 C.I. Basic Red 1 Wool 4
C.I. Acid Green 25 Nylon 5 C.I. Basic Red 14 Acrylic 2
C.I. Acid Orange 3 Wool 5 C.I. Basic Violet 3 Rayon 5
C.I. Acid Orange 7 Acrylic 5 C.I. Basic Violet 10 Wool 5
C.I. Acid Orange 10 Silk 5 C.I. Basic Violet 16 Acrylic 3
C.I. Acid Orange 74 Acrylic 5 C.I. Basic Yellow 11 Acrylic 5
C.I. Acid Orange 156 Wool 5 C.I. Basic Yellow 13 Acrylic 5
C.I. Acid Red 1 Silk 5 C.I. Basic Yellow 15 Nylon 4
C.I. Acid Red 14 Nylon 5 C.I. Direct Black 19 Cotton 3
C.I. Acid Red 18 Wool 5 C.I. Direct Black 22 Cotton 3
C.I. Acid Red 51 Silk 5 C.I. Direct Black 80 Cotton 4
C.I. Acid Red 52 Silk 5 C.I. Direct Blue 71 Cotton 4
C.I. Acid Red 57 Silk 5 C.I. Direct Blue 80 Nylon 5
C.I. Acid Red 87 Wool 4 C.I. Direct Blue 85 Cotton 5
C.I. Acid Red 88 Acetate 5 C.I. Direct Blue 86 Cotton 5
C.I. Acid Red 92 Wool 4 C.I. Direct Brown 1 Cotton 5
C.I. Acid Red 119 Nylon 5 C.I. Direct Green 26 Rayon 3
C.I. Acid Red 127 Acetate 5 C.I. Direct Green 28 Rayon 5
C.I. Acid Red 131 Nylon 5 C.I. Direct Orange 34 Rayon 5
 E. Groves et al. / Forensic Science International 268 (2016) 139–144 143

Table 5 (Continued )

C.I. Fiber type Final extraction value C.I. Fiber type Final extraction value

C.I. Direct Orange 37 Rayon 5 C.I. Disperse Orange 29 Wool 4

C.I. Direct Orange 39 Rayon 4 C.I. Disperse Orange 30 Nylon 5
C.I. Direct Orange 72 Rayon 3 C.I. Disperse Orange 37 Polyester 5
C.I. Direct Orange 102 Nylon 3 C.I. Disperse Orange 44 Cotton 5
C.I. Direct Red 16 Rayon 4 C.I. Disperse Red 1 Nylon 5
C.I. Direct Red 23 Rayon 5 C.I. Disperse Red 1 Rayon 5
C.I. Direct Red 24 Rayon 5 C.I. Disperse Red 5 Acetate 5
C.I. Direct Red 28 Cotton 4 C.I. Disperse Red 11 Nylon 5
C.I. Direct Red 75 Rayon 5 C.I. Disperse Red 17 Rayon 5
C.I. Direct Red 79 Rayon 3 C.I. Disperse Red 50 Polyester 4
C.I. Direct Red 80 Rayon 5 C.I. Disperse Red 54 Polyester 3
C.I. Direct Red 81 Rayon 5 C.I. Disperse Red 60 Nylon 5
C.I. Direct Red 83 Cotton 5 C.I. Disperse Red 73 Polyester 5
C.I. Direct Red 89 Cotton 5 C.I. Disperse Red 82 Polyester 4
C.I. Direct Yellow 11 Cotton 4 C.I. Disperse Red 86 Nylon 3
C.I. Direct Yellow 12 Cotton 5 C.I. Disperse Red 91 Polyester 3
C.I. Direct Yellow 44 Nylon 5 C.I. Disperse Red 92 Polyester 2
C.I. Direct Yellow 106 Cotton 5 C.I. Disperse Red 153 Polyester 5
C.I. Disperse Blue 3 Nylon 5 C.I. Disperse Red 167:1 Acrylic 5
C.I. Disperse Blue 3 Rayon 5 C.I. Disperse Red 167:1 Modacrylic 2
C.I. Disperse Blue 27 Silk 5 C.I. Disperse Red 167:1 Polyester 3
C.I. Disperse Blue 56 Nylon 5 C.I. Disperse Red 167:1 Polyester 5
C.I. Disperse Blue 60 Acetate 5 C.I. Disperse Red 167:1 Rayon 5
C.I. Disperse Blue 73 Nylon 5 C.I. Disperse Red 167:1 Silk 4
C.I. Disperse Blue 79 Polyester 3 C.I. Disperse Red 167:1 Wool 5
C.I. Disperse Blue 102 Acetate 5 C.I. Disperse Red 177 Cotton 3
C.I. Disperse Blue 165 Cotton 5 C.I. Disperse Red 277 Polyester 3
C.I. Disperse Blue 183 Polyester 2 C.I. Disperse Red 311 Cotton 5
C.I. Disperse Blue 284 Cotton 5 C.I. Disperse Red 343 Polyester 3
C.I. Disperse Brown 1 Acetate 5 C.I. Disperse Violet 1 Rayon 5
C.I. Disperse Green 9 Cotton 5 C.I. Disperse Violet 26 Nylon 5
C.I. Disperse Orange 3 Rayon 5 C.I. Disperse Violet 28 Polyester 5
C.I. Disperse Orange 25 Polyester 5 C.I. Disperse Violet 33 Polyester 5
C.I. Disperse Orange 29 Modacrylic 3 C.I. Disperse Yellow 1 Nylon 5
C.I. Disperse Orange 29 Polyester 5 C.I. Disperse Yellow 3 Cotton 5
C.I. Disperse Orange 29 Polyester 5 C.I. Disperse Yellow 3 Rayon 5
C.I. Disperse Orange 29 Rayon 5 C.I. Disperse Yellow 5 Rayon 5
C.I. Disperse Orange 29 Silk 3 C.I. Disperse Yellow 23 Polyester 4
C.I. Disperse Yellow 42 Cotton 5 C.I. Reactive Red 43 Linen 2
C.I. Disperse Yellow 54 Nylon 5 C.I. Reactive Red 43 Nylon 2
C.I. Disperse Yellow 64 Nylon 5 C.I. Reactive Red 49 Rayon 1
C.I. Disperse Yellow 82 Acetate 5 C.I. Reactive Red 120 Cotton 2
C.I. Disperse Yellow 114 Polyester 3 C.I. Reactive Red 141 Cotton 1
C.I. Mordant Black 11 Acetate 5 C.I. Reactive Red 158 Cotton 1
C.I. Mordant Black 11 Cotton 5 C.I. Reactive Red 159 Cotton 1
C.I. Mordant Black 11 Nylon 3 C.I. Reactive Red 180 Rayon 1
C.I. Mordant Black 11 Rayon 5 C.I. Reactive Red 194 Cotton 1
C.I. Mordant Black 11 Silk 3 C.I. Reactive Red 198 Rayon 1
C.I. Mordant Black 11 Wool 2 C.I. Reactive Violet 5 Cotton 1
C.I. Reactive Black 5 Acetate 1 C.I. Reactive Yellow 15 Rayon 1
C.I. Reactive Black 5 Cotton 1 C.I. Reactive Yellow 17 Cotton 1
C.I. Reactive Black 31 Rayon 1 C.I. Reactive Yellow 37 Rayon 3
C.I. Reactive Blue 21 Rayon 2 C.I. Reactive Yellow 84 Cotton 1
C.I. Reactive Blue 160 Cotton 1 C.I. Reactive Yellow 135 Cotton 1
C.I. Reactive Blue 171 Nylon 1 C.I. Reactive Yellow 145 Cotton 1
C.I. Reactive Blue 198 Cotton 1 C.I. Reactive Yellow 160 Rayon 2
C.I. Reactive Green 19 Cotton 2 C.I. Vat Red 13 Cotton 2
C.I. Reactive Orange 16 Cotton 1 C.I. Vat Violet 1 Cotton 2
C.I. Reactive Orange 84 Cotton 1 C.I. Vat Violet 1 Rayon 3
C.I. Reactive Red 43 Cotton 1

159 samples, 82% of the surveyed population, yielded an
extraction value of three or better.
In the course of the evaluations, it was noted that some of the
samples were affected by the extraction solvent. A small number of
rayon and acetate fibers, seven in total, were noted to dissolve in
pyridine:water, but this was not consistent for all rayon or acetate
fibers. Deformation of some modacrylic and acrylic fibers was also
observed. This deformation, observed in only four samples,
resulted in the fibers becoming a brown, congealed mass.
Dissolution or deformation of the fiber is not ideal for forensic
casework; however, dye analyses are usually conducted at the end
of a forensic fiber examination. Furthermore, evaluation of a
known sample (which is typically available in larger quantities
than the questioned sample) prior to analysis of the questioned
sample can serve as an indicator as to the anticipated behavior of
the questioned fiber in the solvent system. If a questioned fiber
does dissolve during extraction, this information provides an
additional point of comparison.
Nine sets of fibers, ranging from two to seven samples per set,
containing the same dyestuff on different fiber types were
evaluated in this study. In general, if the dyestuff was extracted
using pyridine:water as the solvent, it was observed to extract
across all of the fiber types studied, though some variation in
overall extraction ranking was noted. While this only explored a
handful of dyestuffs, it suggests a limited dependence on the fiber
type for extraction efficacy.
144 E. Groves et al. / Forensic Science International 268 (2016) 139–144

Table 6
Summary of pyridine:water extraction results by fiber and dye class. The extraction efficiency (e.g., 1–5) for a given fiber type/dye class combination is the primary number,
while the number of fibers in a particular group is provided in parentheses (e.g., n = 1).

Fiber type

Acetate Acrylic/Modacrylic Nylon PET* Rayon Cotton Wool Silk Linen

Dye application Acid 5 (n = 2) 5 (n = 6) 5 (n = 14) 5 (n = 2) 5 (n = 3) 5 (n = 19) 5 (n = 8)

class
Basic 4 (n = 6) 4 (n = 1) 5 (n = 2) 5 (n = 2)
Direct 4 (n = 3) 4 (n = 13) 4 (n = 13)
Disperse 5 (n = 5) 3 (n = 3) 5 (n = 12) 4 (n = 21) 5 (n = 9) 5 (n = 8) 5 (n = 2) 4 (n = 3)
Mordant 5 (n = 1) 3 (n = 1) 5 (n = 1) 5 (n = 1) 2 (n = 1) 3 (n = 1)
Reactive 1 (n = 1) 2 (n = 2) 2 (n = 8) 1 (n = 17) 2 (n = 1)
Vat 3 (n = 1) 2 (n = 2)
# of samples 9 15 33 21 37 41 24 12 1
*
PET = polyester terephthalate.

Little, if any extraction, was observed for reactive dyes in the
pyridine:water (4:3) extraction solvent. This extraction solvent is
generally not strong enough to break the covalently bound reactive
dyes from the fiber matrix, which likely accounts for the reasons it
is not well represented among the suggested solvents for reactive
dyestuffs (Table 2).
For the entire evaluated sample set, Table 6 provides a summary
of extraction results averaged for each fiber type/dye application
class combination. This table shows the general trend in the efficacy
of pyridine:water as an extraction solvent for textile dyestuffs.
4. Conclusion
These results provide a systematic evaluation of the pyridine:-
water (4:3) solvent system for the extraction of broad range of
commercially significant textile dyes. The extraction of 82% of the
195 fiber samples examined (a population which spans nearly
every major class of fiber type and dye application class) indicates
that this solvent system is applicable to an even wider range of
fiber types, dye application classes, and dye chemistries than
previously known. Therefore, this extraction approach could be
considered as a reasonable first step in a forensic dye analysis
procedure directed toward either comparison or identification.
While extraction is simply one step in the overall analysis of a dyed
textile (to be followed by a separation scheme and possibly
microanalysis), the data presented here provide a means by which
the extraction data alone can yield some initial analytical
constraints on the dye or dyes present in a sample. This, in turn,
has value in both comparative and investigative forensic analyses.
Acknowledgments
The authors would like to thank Katelyn Hargrave and
Stephanie Freiwald for their efforts and contributions. This project
was partially supported by grant 2012-DN-BX-K042 awarded by
the National Institute of Justice, Office of Justice Programs, U.S.
Department of Justice. The opinions, findings, and conclusions or
recommendations expressed in the publication/program/exhibi-
tion are those of the author(s) and do not necessarily reflect those
of the Department of Justice.
References
[1] M. Grieve, The role of fibers in forensic science examinations, J. For. Sci. 28 (4)
(1983) 877–887.
[2] F. Fadil, W. McSharry, Extraction and TLC separation of food, drug, and cosmetic
dyes from tablet-coating formulations, J. Pharm. Sci. 68 (1) (1979) 97–98.
[3] P. Buzzini, G. Massonnet, The discrimination of colored acrylic, cotton, and wool
textile fibers using micro-Raman spectroscopy. Part 1: in situ detection and
characterization of dyes, J. For. Sci. 58 (6) (2013) 1593–1600.
[4] K. Beck, D. Hinks, A. Crawford, N. Weisner, Liquid chromatographic and mass
spectrometric analysis of dyes for forensic purposes, AATCC Rev. (Jan/Feb) (2012)
60–65.
[5] A. Stefan, C.R. Dockery, A.A. Nieuwland, S.N. Roberson, B.M. Baguley, J.E. Hendrix,
S.L. Morgan, Forensic analysis of anthraquinone, azo, and metal complex acid dyes
from nylon fibers by micro-extraction and capillary electrophoresis, Anal. Bioanal.
Chem. 394 (7) (2009) 2077–2085.
[6] A. Carey, N. Rodewijk, X. Xu, J. van de Weerd, Identification of dyes on single
textile fibers by HPLC-DAD-MS, Anal. Chem. 85 (23) (2013) 11335–11343.
[7] K.A. Hargrave, S. Palenik, E. Groves, J. Beckert, C.S. Palenik, Capillary Microspec-
tophotometry of Some Selected Dyed Fibers and Hairs, The Microscope 64 (1)
(2016) 3–12.
[8] R. Macrae, K.W. Smalldon, The extraction of dyestuffs from single wool fibers, J.
For. Sci. 24 (1) (1979) 109–117.
[9] I.B. Beattie, R.J. Dudley, K.W. Smalldon, The extraction and classification of eyes on
single nylon, polyacrylonitrile and polyester fibres, J. Soc. Dyers Col. 95 (8) (1979)
295–302.
[10] B. Beattie, H. Roberts, R.J. Dudley, The extraction and classification of dyes from
cellulose acetate fibers, J. For. Sci. Soc. 21 (3) (1981) 233–237.
[11] I.B. Beattie, H.L. Roberts, R.J. Dudley, Thin layer chromatography of dyes extracted
from polyester, nylon, and polyacrylonitrile fibers, For. Sci. Int. 17 (1) (1981)
57–69.
[12] J.M. Home, R.J. Dudley, Revision of the scheme for extraction and classification of
dyes from polyacrylonitrile fibres, J. Soc. Dyers Col. 97 (1) (1981) 17–19.
[13] A. Greene, The Analysis of Dyestuffs and Their Identification in Dyed and Coloured
Materials, Lake-Pigments, Foodstuffs, etc, Charles Griffin & Company, London,
1920.
[14] E. Clayton, The Identification of Dyes on Textile Fibers and Detection of Metals in
Fibrous Materials, Dyes and Organic Pigments, 2nd ed., The Society of Dyers and
Colourists, Bradford, England, 1946.
[15] C.H. Giles, M. Bashir Ahmad, S.D. Dandekar, R.B. McKay, Identification of
the application class of a colorant on a fibre, J. Soc. Dyers Col. 78 (3)
(1962) 125–127.
[16] Anon., AATCC Technical Manual, American Association of Textile Chemists and
Colourists, Research Triangle Park, 1969.
[17] H. Schweppe, Thin-Layer Chromatography. The Analytical Chemistry of Synthetic
Dyes, Wiley, New York, 1977, pp. 23–55.
[18] K. Wiggins, Thin Layer Chromatographic Analysis for Fibre Dyes. Forensic Exami-
nation of Fibres, 2nd ed., Taylor & Francis, London, 1999, pp. 291–310.
[19] J.C. West, Extraction and analysis of disperse dyes on polyester textiles, J.
Chromatogr. 201 (1) (1981) 47–54.
[20] A.W. Hartshorne, D.K. Liang, The dye classification and discrimination of coloured
polypropylene fibers, For. Sci. Int. 25 (2) (1984) 133–141.
[21] S. Wiersema, Fiber Dye Extraction and TLC, CAC Newsletter, March 1984.
[22] J. Cheng, S.O. Wanogho, N.D. Watson, B. Caddy, The extraction and classification of
dyes from cotton fibres using different solvent systems, J. For. Sci. Soc. 31 (1)
(1991) 31–40.
[23] D.K. Liang, R.J. Duddley, A.W. Hartshorne, J.M. Home, R.A. Rickard, D.C. Bennett,
The extraction and classification of dyes from cotton and viscose fibres, For. Sci.
Int. 50 (1) (1991) 23–35.
[24] ASTM E2227-13, Standard Guide for Forensic Examination of Non-Reactive Dyes
in Textile Fibers by Thin-Layer Chromatography, ASTM International, West
Conshohocken, PA, 2013 www.astm.org.
[25] Pigments and Dyes – A Global Strategic Business Report, Global Industry Analysts,
Rep. USA, 2012.
[26] K. Hunger, Industrial Dyes: Chemistry, Properties, Applications, Wiley-VCH,
Weinheim, 2003.