United States
                  Environmental Protection
                  Agency
Risk Reduction Engineeering
Laboratory
Cincinnati, OH 45268
                  Research and Development
EPA/600/S2-89/049  May 1990
SERA         Project Summary
                   The Selection  and
                   Measurement of Physical
                   Properties  for
                   Characterization of Chemical
                   Protective  Clothing Materials
                   Todd R. Carroll and Arthur D. Schwope
                    Chemical  protective clothing
                  (CPC) must possess certain  physical
                  properties if it is to function as an
                  effective barrier to chemicals. The
                  physical characteristics of CPC mat-
                  erials have gone largely  unstudied;
                  most attention has been focussed on
                  chemical resistance. Physical  prop-
                  erty tests have been surveyed for
                  their applicability to CPC materials,
                  and those tests, which appeared to
                  be most pertinent,  were  applied to
                  ten fabrics and three visor materials.
                  From statistical  analysis of the
                  results and experience gained in per-
                  forming the tests, a minimum battery
                  of tests is recommended. The battery
                  contains nine primary test methods
                  that will allow the  measurement of
                  puncture, puncture-propagation tear,
                  burst, abrasion, accelerated aging,
                  and electrostatic charge  accumula-
                  tion for  CPC  fabrics, and abrasion
                  resistance, deviation in line-of-sight,
                  and  impact  for  the CPC  visor
                  materials. Further development of a
                  cut test is recommended before it is
                  added to the battery.
                    This  Project  Summary  was
                  developed by  EPA's Risk Reduction
                  Engineering Laboratory,  Cincinnati,
                  OH,  to announce key findings of the
                  research project  that is  fully
                  documented in a separate report of
                  the  same title (see Project Report
                  ordering information at back).


                  Introduction
                    Along  with engineering controls and
                  safe work practices, protective clothing is
an important means for minimizing or
preventing the contact  of workers  with
potentially  harmful chemicals.  Such
contacts  can occur in settings which
include industrial plants, waste sites, and
uncontrolled spills.  Those persons
responsible  for worker  protection  must
have available and specify the  most
appropriate  clothing for the particular
situation.  Chemical  Protective Clothing
(CPC) selection,  procurement,  and
specification require information on the
potential severity of the chemical
contacts,  the tasks to be performed, the
skill levels of  the  workers,  the
performance characteristics   and
limitations of the protective clothing, and
the  effect of the clothing on worker
performance.
  Much  has been  written on  the
chemical  resistance of protective clothing
materials, and standard test methods
have been  promulgated. Of equal or
perhaps  greater  importance than the
chemical  resistance of  CPC is that the
clothing remain  intact during the  work
assignment. The clothing  must resist
tears, punctures,  cuts, abrasion,  and
other physical stresses. Although scores
of standard tests exist for measuring
physical properties of the materials  from
which clothing is fabricated, there have
been no studies  directed towards
evaluating the applicability of these tests
to CPC. Consequently, there has been no
basis on which to specify either physical
property testing  or  the  minimum
performance in such testing for CPC.
  The purpose  of this study was to
evaluate  published physical property
tests and to recommend a battery of

-------
tests that could be used  in  CPC
specification and  selection.  The study
was  directed towards tests for  materials
used for garments, in contrast to the
materials  used  in the fabrication of
gloves, boots, and respirators. Tests were
sought that are applicable over the broad
range of  material   types,  able to
discriminate the performances of different
materials,  and relatable to field failure
mechanisms.
   Although  this  study  necessarily
produced  quantitative  physical  property
data, no attempt  has  been made to set
minimum  acceptable values for the
results of  any of  the tests.  Minimum
acceptable values are  dependent on the
specific application of the  clothing. For
example,  the  physical  property
requirements for clothing   used in
laboratory  applications   may be
significantly different  from  those for
clothing used during  entry of  confined
spaces that contain unknown chemical
wastes.
   The recommended  battery  of  tests
does, however, provide both users and
manufacturers of CPC with a means to
compare and evaluate the performance of
CPC. This battery is preliminary in
nature, as  it is used and more data are
generated,  the battery  will be  modified
and  expanded. Additionally, use of the
battery and response  from the field will
aid  in the development of  minimum
acceptable performance values.


Procedure

Materials

   Ten fabrics used to fabricate  pants,
jackets,  coveralls,  or  full  body
encapsulating ensembles, and three clear
plastics used to fabricate visors  were
used in this study (Table  1). Garments
fabricated  from these  materials range in
price from  less than $10 to over $3000.


Candidate Test Methods

   Sources of the standard test methods
used in this  study included  American
Society  for  Testing and  Materials
(ASTM),  Federal  Standards, Military
Standards, American  National Standards
Institute (ANSI), National Fire Protection
Association (NFPA), and  the American
Association of Textile Chemists  and
Colorists  (AATCC). Approximately 50
methods having potential applicability to
the  objectives  of  this  study  were
identified,  and 14 were selected for
laboratory examination. Brief synopses of
the 14 methods follow:
   ASTM F23.20.1-Test  Method  for
Resistance  to  Cut (draft 4) —  The
specimen is mounted  on a holder and
pulled by hand  at a nominal rate of 25
cm/min  beneath  a  single-edged,
industrial razor  blade  which has  been
loaded  with a  known  weight.  The
minimum weight  that produces a  cut
completely through the specimen  is
recorded.  Cut  is  detected  by  the
completion  of  an  electrical  circuit
between  the  razor  blade  and  an
aluminum foil placed under the fabric.
   NFPA 1973 Gloves for Structural Fire
Fighters (Section  3-2.7)- Puncture
Resistance Testing — A machined, 2.03-
mm  diameter stainless  steel  stylus,
having a tip radius of 25 mm,  is pushed
through  a  specimen  at  a  rate  of  127
cm/min.  The force required to puncture
the fabric is recorded.
   ASTM D1424-Tear Resistance of
Woven Fabrics  by Falling-Pendulum
(Elmendorf)" Apparatus — A 12 x 12-mm
notch is cut out of the center of one edge
of a 75 x 75-mm specimen. One side of
the specimen is clamped and the other
side  is fixed  to a pendulum  of  known
weight. A 2-cm slit is  cut into the center
of the  notch and  the pendulum  is
released causing the  specimen  to  tear.
The tearing force  is calculated from the
weight of the pendulum and the distance
that the pendulum travels.
   ASTM D2261-Tearing  Strength of
Woven  Fabrics  by the Tongue  (Single
Rip) Method (Constant-Rate-of-Extension
Tensile  Testing  Machine) — A  8.9-cm
long slit is cut length-wise in the center of
a 76 x 203-mm  specimen. The fabric on
each side of the slit is fastened  into the
jaws (180° opposed) of a tensile testing
machine and the jaws are separated at 5
cm/min.  The tear resistance is the force
required to separate the jaws.
   ASTM D2582-Puncture-  Propagation
Tear Resistance of Plastic Film and Thin
Sheeting- A carriage  of known weight
and holding a 0.32-cm diameter, conical
tipped probe is released from a standard
height such that the probe  impacts,  pun-
ctures, and tears the specimen. The tear
length,  carriage  weight,  and  release
height are recorded and used to calculate
the tear resistance.
   ASTM D3884-Abrasion Resistance of
Textile Fabrics (Rotary Platform, Double-
Head Method) — This method  is  also
  •Mention of trade names or commercial products
  does not constitute endorsement or recom-
  mendation for use.
known as the Taber test. The speci
is  rotated  under  two  abradant wh
under  a  specified  load. The
parameters are the coarseness  of
abradant wheels, the arm weight, anc
number of cycles. In this study, an
weight  of  1,000 g  was  usec
combination  with an H22 vitrified wl
Upon completion of  the  abrasioi
permeation  test, ASTM  F739,
performed on each abraded specime
determine the  effect  of  abrasion
chemical resistance.
   ASTM  04157-Abrasion Resistanc
Textile Fabrics  (Oscillatory  Cylit
Method) — This method is also know
the Wyzenbeek  test.  An  abradan
secured to a barrel (i.e., the cylinder)
the specimen  is  secured  at a  kn
tension in a holder above the cylir
The fabric is lowered onto the oscilU
abradant  and  held  there  unde
predetermined  load for a predeterm
number of cycles (or double rubs). In
study the tension was 2.3 kg, the loac
kg, and the abradant was #80 grit s;
paper.  Similar to the  Taber  tes
permeation test was used as the endp
test.
   ASTM D751-Coated Fabrics, Brea
Strength.  Cut Strip Method —  A  2
100-mm specimen is clamped in the i
of  a  tensile  testing machine and pi
apart at a rate of 30 cm/min. The  fi
required to break the specimen and
elongation at break are measured.
   ASTM D751-Coated Fabrics, Burs
Strength — A 2.5-cm diameter spher
pushed at a  rate of 30 cm/min throuj
4.5-cm  diameter specimen  usini
tensile testing  machine.  The fc
required  to  break the  specimen
measured.
   ASTM  G26-Operating Light-Expo:
Apparatus (Xenon-Arc  Type) with
without   Water  for  Exposure
Nonmetallic  Materials — A  specime
subjected  to periodic  exposure
ultraviolet light and rain under condit
of  elevated temperatures and humidi
In  this study,  a four week  expos
period was used; the exposure cycle
51 minutes  at 90+  %RH and  6l
followed by  9 minutes of rain at 3(
Simultaneously  the  specimens  v\
continuously exposed to a 6000W Xe
Arc lamp operating at 4950W. Specim
were observed  for visual changes
tested for changes in breaking strei
and elongation at break as  describe'
ASTM 751.
   NASA  MMA-1985-79-Evaluating Tr
electric Charge Generation and Deca
A 190 x 190-mm specimen is rubbed
a  polytetrafluoroethylene  (PTFE)

-------
overed wheel rotating at 200 rpm under
  1.36  kg  load  for  10 seconds. The
harge (i.e., voltage) on the specimen is
measured  immediately  (i.e.,  the peak
oltage) and 0.5, 1, 2, 3, 4, and 5 sec-
nds thereafter.
  ASTM D881-Deviation of Line-of-SigM
- A 150 x  150-mm specimen of a visor
material is  held in a position normal to a
le-of-sight established between a fixed
slescope   and  target.  The   angular
eviation caused  by the visor material is
etermined from the apparent shift of the
rosshairs of the  telescope, the  distance
etween the specimen and  the target,
id the spacing of the lines on the target.
or this investigation, the target consisted
  vertical  lines  spaced 0.25 cm apart
ith the target set 340 cm from  the
secimen.
  ASTM  01044-Resistance  of Trans-
arent  Plastics  to  Surface  Abrasion
iaze) —  This test is identical to  the
iber  Abrasion  test,  however,  a finer
Dradant  is typically used  at  fewer
/cles. A CS10 abrading wheel was used
  an arm weight of 1,000 grams for 10,
5, and 50  cycles.  The  loss  in light
 msmittance at 550 nm was measured.
  ASTM D3029-lmpact Resistance of
igid Plastic Sheeting — A  5 x 5-cm
secimen is held in an aluminum clamp
 iich has a 3.8-cm diameter hole in it. A
  5-gram dart is  released from  a known
   Table 1. Test Materials
          height (2.5 cm to 175 cm) and allowjed to
          impact the  specimen. The  mean failure
          energy is calculated  from the  weight of
          the dart and height at which 50% cf the
          specimens  failed. A  failure was defined
          as visually (unaided) detectable  craqks in
          the specimen.

          Results and Discussion      :

             The results  for all tests, except  the
          abrasion tests, are summarized  in Tables
          2 and 3. The content and organization of
          Table 2 warrants discussion. Mean values
          for each test of each fabric are reported
          along with  the standard deviation (in
          parentheses). The number of replicates,
          n, is  designated  beneath each column
          heading.  The  upper  case letter  to<  the
          right of the standard deviation is e;ither
          the  Duncan's  or  the  Tukey's   Groubing
          Letter (DGL or TGL, respectively). Vyithin
          each  test,  the  letters designate  results
          that are statistically similar or dissimilar at
          the  95% confidence  level.  For example
          under puncture,  the  result  for I the
          supported  butyl fabric  (DGL  =  A) is
          significantly different from  that  ofi the
          supported PVC (DGL =  B) but  the result
          for the supported  CPE (OGL =  C) is: not
          significantly different from  that for; the
          Viton-Nomex-Chlorobutyl (DGL = Q.I
             Table 3  for the  visor  material^ is
          organized  in a similar manner with- the
                          exception that no statistical analysis was
                          performed.
                            Cut — Cut resistances of the fabrics
                          ranged  from  365  g to  1265  g. The
                          method was  easy  to perform and  the
                          apparatus relatively inexpensive to build.
                          The draft method, however, has  several
                          shortcomings which must be corrected
                          before the method can be considered as
                          part of  a  standard test battery. These
                          shortcomings  include:  lack  of  a
                          standardized  industrial  razor blade,  the
                          absence of a means for controlling the
                          rate at which the fabric  holder is pulled
                          under the razor  blade,  the  large weight
                          increments  that prevent differentiation of
                          some  fabrics, and the  means by which
                          test results  are  generated  that renders
                          them  difficult to  analyze by  common
                          statistical  methods. This  latter
                          shortcoming  prevented  analysis of the
                          data by Duncan's  multiple range test;
                          thus the DGLs are absent from  Table 2.
                          Questions exist as to the applicability of
                          the  method to all garment materials and
                          the  relationship of  the test  to  field
                          scenarios.
                            Puncture —  The  resistance of the
                          fabrics  to puncture ranged  from  3.3 to
                          19.4 kgf. The puncture test  has good
                          precision; the relative standard deviation
                          was  less  than  10% for each  fabric.
                          Whether this test is representative of field
                          puncture scenarios is open to discussion.
             Material
        Description
            Source
Weight.'
 glm2
                                                                                                        Thickness,
                                                                                                            mm
    Fabrics
    flirty/ Rubber (supported)
    Challenge 5200®
    Chemrel®
    CPE
    CPE (sup
    PE-Tyvek®
    PVC (supported)

    Saranex9- Tyvek
    Viton9-Nylon- Chlorobutyl
    Viton-Nomex9-Chlorobutyl
Butyl -Nylon Fabric -Butyl
Tefton-Nomex-Teflon
Multilayer Plastic Film-Fabric
Chlorinated Polyethylene
CPE-Polyester Fabric-CPE
Polyethylene-Tvvek
Polyvinyl-Chloride-Polyester
Fabric
Saranex-Tyvek
Viton-Nylon Fabric-Chlorobutyl
Viton-Nomex-Chlorobutyl
Fyrepel Products Inc.
Chemical Fabrics Corp.
Chemron, Inc.
ILC Dover
Standard Safety Equipment Co.
Kappler, Inc.
Standard Safety Equipment

Kappler, Inc.
Life-Guard, Inc.
Fairprene, Inc.
  428
  528
  145
  698
  743
  76
  898

  126
  584
  683
0.37
0.26
0.27
0.52
0.60
0.14
0.76

0.19
0.44
0.42
    Visor Materials (flat)
    FEP (film)
Fluorinated ethylene
propylene copolymer
Chemical Fabrics Corp.
  -t
0.25
    Polycarbonate
        Melamine-coated
        Uncoafed
    PVC (flexible)
Polyvinyl chloride
General Electric

Sheffield
Standard Safety Equipment Co.
                1.02

                0.76
                1.02
     "Average of 5 measurements.
     +Average of 20 measurements.
     /•A/of measured.

-------
V.
£
co
!
«*«
S*i
£
o
o
o
£

If
114
*i

5?
cf ^
O m
««
O) C
5 I"
w rr
c "J
.£
c o -
l°*
"• o:

"co °»
s. - s"?
1 . f
t;
CO
« 0,_
* H~-CN
f IS
1 in
CO S v-
V Gj
1 .
o ^
w $«
c a "
ID O C
0 C v-
§ £
Q
T3
C
CO
S 67??
1 31
CO
CD
Table 2. Fabrics T
Material



I5907(2164)A
,




329(16)A

R
^* •*-*
• CD
CO ^
^r°o
CO T*
co '
CO

CM"
N. •<-
,; _ CD
co00"
CM '
00

§
^~
o^
CM"
oo

5605(375)B


•»•


5^
co"
CO
C3^
i;
C)


5
io"
CO
CO

Butyl Rubber
(supported



7847(1330)A
"




143(28)D

0
T^CM
S.S
O) '
Is!

i^
°? 0>
2-
V

s.
^ 10
stg
CD "•>
•<»'
T

20.8(0.69)
F
-76.1

O
co
CM
cj
CO
CO

0(981)688


422(40)D


1
o
^
^


o]
•o"
CO
^

Chemrel



|
V





£
s

s.
*^ C\
sS
10 rJ
IS.
CM

20.3(0.97)
F 7.2

•^
^_
^-~
CO
c\

843(94)G


'


ua^
?
CM
cj^
o>
*^
o»


3-
S
CO

s



a(zs9)eoti.





UJ^
?"
S

g
o
^ <°.
^t^-
10

^
5^2
is!
CO

O
CO
TJ.

O^
10
T~

1
CO
Sr
CO
10
CO


1
CM
CO


^
^f
O)
c^
IS.
in


o
8
>a-

CPE(supported)



Q
I
V





ft
CM
10

o
t?o
»M .
r-' 0
J=rO
O T-
IO

O
2q
oJ2
m'

£
CO
CO
d
Sr
CO

1 360(1 73)F


'•


£
IO
10
o
co
CO


o
io"
CO
CO

I
i



1
V





1
O)

<*
00
QCM
^ *"
CO
»*

5-
^00
^•"7
CM

OQ
S1

s^
(S^


2340(1 35)D,E


•


s.
in"
ccT
is!


5
!o"
CO
CO

PVC(supported)



Q
|
V





CD
fc"

O
O"*" QQ
«AJ
r^ C3)
— in
CM •
00

6.8(0. 32 )G
-48.3

^

^_
CD
'j1

1633(299)F


216(369)0
*-

tts
%
csT
•^


§
«
%

Saranex-Tyvek



97(3680)8,0
CO
^^



O
CN~
I

S-

CO *
5^ co
"O
CO

C7?
CM,. <9
sT° °?
CD
CO

at
CM
CO
o
ar
in

2059(41 )E


1109(33)0


8-
00
o
c\r
CO
f~.


o"
8
CO

Viton-Nylon-
Chorobutyl



•457(116)0





93.
in
1

QQ
<^
I
1
§
^.
1
s
1
t
CD
Q.
S
's analysis w
«•
J

!R
§
«
0)
'ukey's Groupin
o specimens.
r^. >
k» ^
Duncan's o
average of
^ 9
i I
BJ *0
S CO
* 1
* i
§ 3
S e
« Q
5 1
CD 0
5 ^
% a
•o S
CD Q.
S C
' Results are repc
+ Not tested.
t Percent change i,

-------
3uncture  is in part  determined by the
speed at which an object impacts a fabric
md the freedom that  the fabric  has  to
ilongate  upon the  impact.  These
parameters have not  been  studied  in
lither the laboratory or the field.
  fear — This  study included  the
nvestigation  of  three  tear  tests:
Elmendorf,  tongue,  and  puncture-
Kopagation.
  Both the Elmendorf and the tongue
jar tests begin with a fabric specimen
mich has been slit. Thus, these tests do
ot measure the resistance of the fabric
3 tear initiation, rather, they measure
nly the resistance to  tear propagation.
 ince  the CPC  issued to workers  is
resumed free of cuts, tears, holes, and
o forth,  these tests may  not fully
 ^present field  failure mechanisms.
 urthermore, the Elmendorf test yields
 nly  the  maximum value of  the  tear
 isistance for a fabric and the  tear
 attern  exhibited  by the  nonwoven
 jecimens was  not  consistent with the
 iquirements of the method. The types of
 'suits produced  by the Elmendorf
 jparatus varied with the type of fabric
 jpport (woven vs.  nonwoven) and  may
ot be comparable. The values in Table 2
 re mean tearing forces.
  More representative  of a field tear, the
uncture-propagation  tear  simulates a
 edition of snag. The force required  to
 itiate and propagate a tear is measured.
 lis test was applicable to all fabrics and
 ! precision, as judged by a comparison
   relative standard  deviations,  was
 jnsiderably greater than those  of the
 Imendorf and tongue tear tests.
  Abrasion  —  Several  abrasion
rocedures were  investigated  in an
 tempt to identify a method  that would
 ) representative of field conditions.  In
 is  study, ASTM  F739 was  used  to
 easure the effect  of abrasion on the
 temical  resistance  of  the  fabric.
 cetone was  used as the  chemical
  allenge for  all permeation  tests.  The
  ect  of  abrasion  on  a material  was
 dged  by the change in breakthrough
  le of the acetone.
  Abrasion  testing is  generally
 cognized as semi-quantitative  in
haracter;  reproducibility is difficult  to
  hieve. The  actual  abrading  action on
he  fabric  is dependent  on  the
oarseness  of the abradant,  the  weight
pplied to the  abradant, and the number
f abrasion cycles as well as the tautness
f the fabric. Fabric tautness may change
 uring the procedure as the fabric heats
 id stretches  due  to the  abrasion
 •ocess and may vary  from fabric  to
 brie.  A general rule of thumb is that the
reproducibility of the abrasion  increases
as the number of cycles is increased and
as the coarseness  of the abradant  is
decreased.
   For  the  Wyzenbeek  abrader,  the
breakthrough times  for acetone and the
Viton-Nylon-Chlorobutyl fabric  remained
relatively stable for  25, 50, 75, and  100
cycles, then dropped precipitously at 250
cycles. Even 25 cycles was sufficient to
cause immediate  breakthrough  of
acetone  through  the Saranex-Tyvek
material.  Breakthrough  of  acetone
through Challenge 5200 remained above
two hours even after 400 cycles.
   Midway   through   the   study
consideration of the  Wyzenbeek test was
discontinued in favor of the Taber test for
three reasons. One,  the Taber abrasion
pattern  is  more  uniform. Two,  the
Wyzenbeek apparatus  is no  longer
commercially available. Three, the lack of
its commercial availability would seem to
suggest that the Wyzenbeek method has
previously not been found useful by the
textile fabric test community. The Taber
test, on  the other  hand, is  widely
performed  and the  apparatus  is readily
available.
   Five hundred Taber cycles  caused
immediate  breakthrough  of  acetone
through  the Chemrel fabric  but  had
minimal (if any) effect on the supported
CPE and  no apparent effect  on  the
Challenge 5200.
   Abrasion testing  with  a permeation
endpoint  test  was  successfully used to
discriminate the performances of  CPC
fabrics. This study, however, has  not
resolved  precision shortcomings that are
characteristic of abrasion testing nor have
test conditions (e.g.,  abradant coarseness
and  load) been  defined that  represent
field  scenarios.
   Tensile  Strength  and  Elongation at
Break — These common  tensile tests
were applied to new  fabrics and to fabrics
that  had  been subject to accelerated
aging.  The  results  are summarized in
Table 2  in a  format that has  the initial
values on  one  line  and  the percent
change in the values due to the aging
immediately underneath.   These
measurements are easily performed with
equipment that is commercially available.
Good precision was found and the tests
were applicable to all fabrics.
   Static  Charge Accumulation — As is
evident  from  Table  2,   the fabrics
exhibited a wide range of abilities to hold
and  dissipate voltages  produced by
rubbing the fabrics with a PTFE wheel. In
reviewing the data,  one  must bear in
mind that  the absolute  value  of  the
reported  result, not its sign, is important.
This characteristic should be considered
when selecting or specifying CPC since
static charges could lead to sparking with
disastrous  consequences in certain
situations involving chemicals.
   This method is applicable to all types
of CPC  materials and appears to  enable
discrimination of the results.
   Deviation in Une-of-Sight — This test
is designed to measure the deviation in
the line-of-sight caused by a clear plastic.
To be  useful,  the  results  must  be
obtained  with  the plastic in its use
configuration; in  the  case  of  visors, this
typically means curved. If the deviation in
line-of- sight becomes noticeable, users
of the visor materials will  have difficulty
manipulating objects  which  they are
focusing on.
   Table 3 summarizes the limited  results
generated during this investigation. Of the
four materials tested, only  the PVC
caused any measurable deviation in the
line-of-sight. The method  is  easy  to
perform and  appears to be applicable to
all visor materials.
   Waze — The  Taber test  conditions
used  in this study  were   arbitrary but
provide comparative data on the abrasion
resistance  of  visor materials.  Visor
abrasion can occur  during  suit use,
decontamination and storage.
   Table  3  summarizes the measured
decrease  in light transmission. The
Melamine-coated  polycarbonate retained
the highest amount of light transmission
at 50 cycles. The results  show  good
precision and it  is apparent from the
differences in the results that the method
can discriminate among the performance
of different visor materials.
   This test is quick and easy to perform
but requires the use of a  spectro-
photometer. The only apparent limitation
to this method is that it can only be used
on flat specimens.
   Visor Impact — The test simulates the
impact of a  small diameter,  semi-sharp
projectile with a visor. Direct relation  of
the results to the field, however, is com-
plicated by the fact that the impact during
testing can only occur within the diameter
of the specimen holder. The  size of the
holder restricts the potential for deflection
and flexing in the specimen.
   Table 3 summarizes the  results of the
impact test for the four materials tested.
The FEP was the only material made  to
fail by this method. The limitations of our
apparatus  were a maximum drop  height
of 175 cm and a maximum dart weight of
315 gram.
   The test  is easy  to  perform and
applicable  to all types  of flat  visor
materials.  Some ambiguity  exists  in

-------
selecting  the  correct  number  of test
specimens. The  endpoint of the test is
reached when 50% of the specimens fail.
Statistical Analysis
   As discussed  above, the results from
each of the fabric tests were subjected to
either a Duncan's or Tukey's analysis in
order to demonstrate the degree to which
the test could be used to discriminate
between the performances of the fabrics.
Another objective of  the  study  was  to
identify  and eliminate those tests that
seemed to be redundant.  A minimum
battery of tests was desired that  would
provide  broad   perspective on the
physical  characteristics  of protective
clothing materials.
   Test  method redundancy  was
investigated by applying the Spearmen's
Rank Correlation  Coefficient (SRCC)
procedure.  If the  rank orders of the two
lists were exactly the same the coefficient
would be 1, if the rankings were exactly
opposite then the coefficient would be -1.
A  coefficient greater than  0.65  is
indicative of a pair of lists in which the
rank  orders are  in  relatively  good
agreement.
   Strong  correlations  were  found
between the rank orders of the fabrics for
the burst  strength/break strength pair
(SRCC = 0.92)  and the burst strength/
puncture strength pair (SRCC = 0.84).
Good correlations were found  between
the rank orders  of the puncture  resis-
tance/tongue tear pair (SRCC = 0.70) and
the puncture strength/break strength pair
(SRCC=0.72),  and between  the  tongue
                            tear/burst strength pair (SRCC = 0.67) and
                            the  tongue  tear/break  strength  pair
                            (SRCC = 0.65).
                            Conclusions and
                            Recommendations

                            Recommended Test Battery
                               The standard tests and conditioning
                            methods  listed in  Table 4 are  recom-
                            mended as the minimum battery of pro-
                            cedures for  characterizing or specifying
                            the physical properties of  chemical pro-
                            tective clothing materials. These stan-
                            dards can be supplemented  with others,
                            dependent on the needs  of each specific
                            application of the clothing.
                               The puncture-propagation  tear test
                            was  selected  over  the  Elmendorf and
                            tongue tear  tests because it appears to
                            yield unambiguous results for all fabrics
                            and because it appears  to most closely
                            represent field  failure mechanisms for
                            garment materials.
                               The burst test is recommended over
                            the more commonly performed tensile
                            test because the burst  test is easy to
                            perform  and  is not subject  to  the
                            confounding problems  of jaw breaks,
                            fabric  slippage, or fabric orientation.
                            Furthermore,  from the  Spearmen's
                            analysis, the tensile test appears to  be
                            redundant of the burst test.
                               The Taber abrasion test was selected
                            over the  Wyzenbeek test because of
                            availability problems of  the  Wyzenbeek
                            apparatus  and because  it can produce
                            specimens of suitable size  and quality for
                            endpoint testing. The permeation test is
recommended as the endpoint test
assessing  the  effects  of  abrasion
chemical  resistance.  Although i
investigated in this  study, the effects
abrasion  on the  physical integrity of
fabric could be  measured  by the bi
test.
   Accelerated  aging  followed  by
same endpoint tests as for the abra:
test  is recommended for those clotf
use  scenarios that  include a  signific
amount of  reuse or extensive stor
periods.
   The triboelectric  charge  test has
ticular applicability  to  work  scena
involving  flammable or explosive r
erials. Such clothing  should  also
subject to  flammability  or  flame re
tance testing.  Flammability testing
beyond  the  scope  of  this  study
consequently there  is  no  specific
recommendation herein; other refere
should be consulted.
   Although the authors believe a cul
should be part of a minimum test bal
none is included because the pres
available methods  have  been  juc
inadequate. Further  work in this ar
recommended.  Further work is
recommended pertinent to the abr
test. Finally, efforts should be  under
to develop  case history  files of
failures due to tear, cut, puncture,
and  so forth in order to establish  a
base for minimum performance spec
tions for each of the  tests.
   The full report  was submits
fulfillment of Contract No. 68-03-32
Arthur  D. Little,  Inc.,  under
sponsorship of the  U.S. Environn
Protection Agency.
            Table 3. Visor Materials Test Results
Material
FEP (film)
Deviation
Thick- of line of '
ness'.cm sight, * ,min
0.025 -T
Haze, % transmission @ 550 nm abrasion cycles
0
89(1 r
(i/3)tt
10
72(3)
(3/3)
25
72(2)
(313)
50
67(2)
(3/3)

Impact
strength, i
1.98 + +
(22)~
             Polycarbonate
Melamine-
coated
Uncoated
PVC (flexible)
0.102
0.076
0.102
0
0
3.1
88(1)
(1/3)
88(3)
(1/3)
85(1)
(113)
84(6)
(1f3)
72(2)
(2/3)
73(3)
(313)
33(2)
(1/3)
62(2)
(2/3)
62(3)
(3/3)
73(2)
(1/3)
58(2)
(2/3)
53(4)
(3/3)
>5.43
(10)
>5.43
(10)
>5.43
(10)
ft
                  Average of 20 measurements.
                 Maximum deviation measured in three specimens.
                  Not tested.
                 Average % transmission (standard deviation).
                 Average impact strength.
                  Number of specimens tested/number of measurements made on each specimen tested.
                Number of specimens tested.

-------
Table 4.     Recommended Physical Property Test Methods For Chemical
            Protective Clothing
 Fabrics
 Puncture resistance
 Puncture-propagation tear
 resistance
 Abrasion resistance endpoint tests:
 Bursting resistance
 Accelerated aging endpoint tests:

 Electrostatic charge
 Visor Materials
 Deviation in line-of-sight
 Haze-abrasion resistance
 Impact resistance
NFPA 1973- Paragraph 3-2.7
ASTM 02582
ASTM 03884
Permeation-ASTM F739
Burst Strength-ASTM D 751
ASTM 751
ASTM G26
Permeation-ASTM F739
Burst Strength- ASTM 0751
NASA MMA-1985-79
ASTM 0881
ASTM 01044
ASTM 03029

-------
  Todd R. Carroll and Arthur D. Schwope are with Arthur D. Little, Inc., Cambridge,
  MA 02140-2390
  Michael D. Royer is the EPA Project Officer (see below).
  The complete report,  entitled "The Selection and Measurement  of Physical
        Properties for Characterization of Chemical Protective Clothing Materials,"
        (Order  No. PB90-188-731/AS; Cost:  $17.00  subject to change)  will  be
        available only from:
            National Technical Information Service
            5285 Port Royal Road
            Springfield, VA 22161
            Telephone: 703-487-4650
  The EPA Project Officer can be contacted at:
            Superfund Technology Demonstration Division
            Risk Reduction Engineering Laboratory
            Edison, New Jersey 08837-3679
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
                                                                                            -
  ^OFFICIAL
PENALTY
                                                                                                     0  35
Official Business
Penalty for Private Use $300

EPA/600/S2-89/049
       000085833    PS
       230 S  0EA8BOH8  STREET
       CHICAGO

-------