United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
Research and Development
EPA-600/S4-84-071 Sept. 1984
^EPA Project Summary
Evaluation of Turbidimetric
Methods for Monitoring of
Asbestos Fibers in Water
E. J. Chatfield, M. J. Dillon, and W. R. Stott
A rapid and inexpensive technique is
needed for the routine monitoring of
drinking water and other water supplies
to reveal asbestos fibers. To this end,
the use of turbidity measurements
obtained with a commercially available
but modified turbidimeter was investi-
gated in a 1-year study.
The modified turbidimetric tech-
nique, which uses magnetic separation
for the prior concentration of fibers,
allowed monitoring of high-iron amphi-
boles at the one-million-fiber-per-liter
level. Additional instrument modifica-
tion and improvements in the method-
ology are needed to permit monitoring
of low-iron amphiboles and chrysotile
at environmentally significant concen-
trations.
This Project Summary was developed
by EPA's Environmental Research Lab-
oratory. Athens, GA. to announce key
findings of the research project that is
fully documented in a separate report of
the same title (see Project Report order-
ing information at back).
Introduction
The analytical method for determina-
tion of asbestos in water samples has
been progressively improved since the
Preliminary Interim Method was issued
in 1976 by the U.S. Environmental
Protection Agency's (EPA's) Environ-
mental Research Laboratory in Athens,
GA. The method is based on filtration of
the water sample through a sub-
micrometer pore diameter membrane
filter, followed by preparation of the filter
for examination in a transmission elec-
tron microscope (TEM). Individual fibers
are then identified, measured, and
counted.
Asbestos analyses using the TEM
method are time-consuming and
expensive because each sample requires
examination in an analytical electron
microscope for at least 3 hours, during
which every fiber is separately classified
and measured. Satisfactory analyses
require that the microscope operators
have a high degree of skill and experience.
The high cost and the long turnaround
time of analyses made by electron
microscopy have prompted investigations
into whether more rapid analytical
procedures could be developed, even if
these were not capable of comparable
specificity and sensitivity. Several
potential technical approaches based on
optical scattering by fibers have been
investigated recently. These instrument-
al techniques, however, are at the
prototype stage. Accordingly, there is
interest in the concept of using
established instrumental techniques,
perhaps with minor modifications, to
monitor asbestos fiber concentrations in
water.
Turbidimetry is a sensitive technique
that has been applied by several investi-
gators to the problem of measurement of
asbestos in drinking water. In these in-
vestigations, however, it appears that the
following two apparently contradictory
conclusions were reached:
e Turbidity readings can be used as an
indicator of fiber removal efficiency
in a filtration plant.
e No definite correlation between fiber
concentration and turbidity was
found.
-------
The situation is complicated further by
the fact that most of the fiber concentra-
tion values reported in these studies of
turbidity would now be considered
questionable because of the new
methods used in the preparation of
specimens for the TEM evaluations.
Clearly, there is no reason why the
asbestos fiber concentration should be
related to the turbidity measurement,
unless the asbestos is the principal
particle species that gives rise to the
turbidity in the particular water sample.
This is unlikely to be-the case in most
water systems. Nevertheless, turbidity
measurements may be useful in
monitoring asbestos fiber concentrations
in two ways: first, if the fibers could be
concentrated selectively so that they form
the major contribution to the turbidity
measurement; and secondly, if the
turbidity of the sample could be reduced
so that the residual value corresponds to
a fiber concentration below some level of
concern. There also would be an
advantage if the turbidity measurement
itself could discriminate between the
contributions made by fibrous a ndequant
particles. In the study reported here, all of
these aspects were investigated to deter-
mine whether measurement of turbidity
could provide a realistic method for
monitoring asbestos fiber concentrations
in drinking water, with TEM reserved for
more specific characterization.
Conclusions and
Recommendations
The experiments showed that the
Sigrist L65 turbidimeter was capable of
detecting a turbidity increment of 0.002
nephelometric turbidity unit (NTU) above
the value for non-boiling still water.
Other turbidimeters were not able to
achieve this detection level.
Relationships between turbidity and
asbestos fiber concentration were
obtained for UICC crocidolite, UICC
amosite, and Union Carbide refined
Calidria chrysotile. These indicated that a
contribution of 0.004 NTU to the total
turbidity corresponded to asbestos fiber
concentrations of 5 million fibers per liter
(MFL), 2 MFL, and 1000 MFL for UICC
crocidolite, UICC amosite, and Union
Carbide chrysotile, respectively. These
corresponded to mass concentrations of
about 1.3pg/L, 1.3A/g/L, and 6.0/jg/L,
respectively, for the three varieties of
asbestos.
A modification was made to the turbidi-
meter that allowed the sample to be
placed between the poles of an
electromagnet. The magnetic field
caused the fibers to become oriented, and
it was possible to observe a change in the
turbidity reading when the field was
applied.
A change of 0.004 NTU in the turbidity
reading corresponded to fiber concentra-
tions of 3.5 MFL, 12 MFL, and 20,000
MFL for UICC crocidolite, UICC amosite,
and Union Carbide chrysotile,
respectively. These corresponded to
mass concentrations of 0.8 pg/L, 7.5
fjg/L, and 100 /jg/L, respectively,for the
three varieties of asbestos.
In the modified turbidimeter, with
application of the magnetic field, fibers
that adopted an orientation parallel with
the field (p-fibers) caused an increase in
the turbidity reading; fibers that adopted
orientations normal to the field (n-fibers)
caused a reduction in the turbidity
reading. Increases in the turbidity
readings were obtained using crocidolite,
amosite, chrysotile, anthophyllite, and
cummingtonite. Decreases were observed
using tremolite and actinolite. In the case
of amosite, the signals from the two
alignment modes of fiber partially
canceled each other, and different magni-
tudes of effect were obtained with two
different varieties of amosite. Theoretic-
ally, some fibrous minerals or mixtures
could exist in which the effects from the
two alignment modes of fiber could
completely cancel each other, leading to
an incorrect conclusion that no fibers are
present.
The detection levels of the modified
turbidimeter system could be improved by
use of more sophisticated time-averaging
and read-out techniques.
Magnetic separation was an effective
procedure for concentration of crocidolite
from water samples. Starting with 4-liter
water samples, the amphibole fibers
could be separated and transferred to a
65 mL water sample for the turbidimeter.
The numerical fiber recovery was
variable for some reason, between 14%
and 43%, although the mass recovery
was constant at about 65%, indicating
that the larger fibers were recovered
more reproducibly. Concentration of the
fibers by factors of between 8.7 and 26.9
were achieved, equivalent to a factor of
40 in terms of mass concentration.
Magnetic separation cannot be used for
concentration of chrysotile and low-iron
amphiboles, and other methods must be
used.
To improve detection limits, techniques
were investigated to reduce the turbidity
of the water sample while retaining the
majority of the asbestos fibers. Filtration
of the sample through large pore
diameter Nuclepore filters was studied as
a means of separating the asbestos fibers
from other paniculate material on the
basis of differences in their size distribu-
tions. For a typical size distribution of
waterborne chrysotile fibers, Nuclepore
filters of pore diameters exceeding 0.8
fjm allowed most of the chrysotile asbes-
tos fibers of an artificial dispersion to pass
through. This was demonstrated using
0.8/vm, 2.0/vm, andS.Opm porediameter
filters. After such a filtration, the
turbidities of typical drinking water
samples were significantly reduced,
usually by a factor of about 10 when 0.8
fjm pore diameter filters were used.
The technique was more effective for
samples of drinking water if the organic
materials were oxidized prior to this
selective filtration step, indicating that
many of the fibers were associated with
organic materials in the original drinking
water sample. Using a 2.0 (im pore
diameter filter, 78% of the fibers in an
artificial dispersion prepared using a
drinking water sample were found to pass
through the filtrate, and, at the same
time, the turbidity was reduced from
0.109 NTU to 0.0165 NTU. Therefore, this
procedure would achieve an improvement
of about a factor of 5 in the detection limit.
The turbidity measurement, made by
the modified instrument, with prior con-
centration of the fibers by magnetic
separation, would allow fiber-specific
monitoring of high-iron amphiboles at the
1 MFL level. The utility of the instrument
for low-iron amphiboles depends on the
extent to which the detection level can be
improved by signal averaging and on the
effectiveness of the selective filtration
technique for amphibole fibers.
Additional modification of the instrument
would be required before it could be used
to monitor chrysotile at realistic concen-
trations.
-------
E. J. Chat field, M. J. Dillon, and W. R. Stott are with Ontario Research Foundation.
Mississauga, Ontario, Canada L5K 1B3.
J. M. Long is the EPA Project Officer (see below).
The complete report, entitled "Evaluation of Turbidimetric Methods for Monitoring
of Asbestos Fibers in Water," (Order No. PB 84-232 51 J; Cost: $11.50, subject
to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
Athens. GA 30613
•& U S. GOVERNMENT PRINTING OFFICE; 1984 — 759-015/7817
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use S300
------- |