WATER POLLUTION CONTROL RESEARCH SERIES • 16080 HUB 04/72
WASTE WOOL AS A SCAVENGER
FOR MERCURY POLLUTION IN WATERS
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our Nation's waters. They provide a central source of
information on the research, development, and demonstration
activities in the water research program of the Environmental
Protection Agency, through inhouse research and grants and
contracts with Federal, State, and local agencies, research
institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, DC 20460.
-------
WASTE WOOL AS A SCAVENGER FOR
MERCURY POLLUTION IN WATERS
by
Joseph P. Tratnyek
Arthur D. Little, Inc.
Cambridge, Massachusetts 02140
for the
Office of Research and Monitoring
Environmental Protection Agency
Project No. 16080 HUB
Contract No. 68-01-0090
April 1972
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C., 20402 - Price 60 cents
-------
EPA Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication.
Approval does not signify that the contents necessarily
reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or
recommendation for use.
ii
-------
ABSTRACT
A laboratory study was conducted with a variety of available industrial
waste fibers of wool, wool/polyester, and nylon to determine the feasi-
bility of using such fibers to remove mercury from waters and bottom
deposits contaminated with mercury. Inorganic and organic sources of
mercury were utilized and parameters affecting mercury removal were
investigated.
Mercury removed by wool fiber amounted to 90-95% in 24 hours at the
1-ppm level used in the majority of our experiments. At higher levels
of mercury, larger quantities were removed up to 300 mg Hg/gram of
fiber, but the percentage decreased for a given amount of fiber. Changes
in pH (2 to 10) and temperature (5 to 35°C) did not markedly alter
efficacy of wool, nor did anaerobic conditions or variation in water
hardness. The presence of sulfide in water or sludge reduced effective-
ness. Nylon has limited potential as a scavenger for mercury, removing
20-50% of the mercury depending upon circumstances. Waste wool appears
to be a potentially useful material for removing mercury from contami-
nated waters.
This report was submitted in fulfillment of Project Number 16080 HUB,
Contract Number 68-01-0090, under the sponsorship of the Office of
Research and Monitoring, Environmental Protection Agency.
iii
-------
CONTENTS
Section
I
II
III
IV
V
VI
VII
VIII
IX
X
Conclusions
Recommenda t ions
Introduction
Description of Fibers
Sorption of Mercury by Fibers
Factors Relating to Mercury Sorption
Effect of pH
Effect of Temperature
Capacity for Mercury
Quality of Water
Anaerobic Condition
Effect of Washing Fiber
Mercury Removal from Sediment
Discussion
Mechanism of Sorption
Field Utilization and Cost
Acknowledgements
Appendices
Page
1
3
5
9
11
21
21
23
25
26
27
28
31
33
33
34
37
39
-------
FIGURES
Page
1 Depletion of Mercury from Mercuric Chloride Solution ^
2 Depletion of Mercury from Methyl Mercuric Chloride Solution 18
VI
-------
TABLES
No. Page
1 Fibers Tested 9
2 Mercury Content of Selected Fibers 10
3 Material Balance 12
4 Depletion of Mercury (50 ppm) from Solutions Containing 13
Fibers
5 Depletion of Mercury (1 ppm) from Solution Containing 14
Mercuric Chloride
6 Depletion of Mercury (1 ppm) from Solution Containing 15
Methyl Mercuric Chloride
7 Depletion of Mercury (1 ppm) from Solution Containing 16
Phenyl Mercuric Acetate
8 Depletion of Mercury from Solutions Containing 19
Bis (2-Methoxyethyl) Mercury and Dissolved Metallic
Mercury
9 Depletion of Mercury at Various pH 22
10 pH of Unbuffered Solutions Containing Fibers 23
11 Depletion of Mercury at Several Temperatures 24
12 Depletion of Mercury at Various Concentrations of 25
Mercuric Chloride
13 Depletion of Mercury by Fibers with Repeated Contact 26
with 1 yg/ml Mercuric Chloride Solution
14 Depletion of Mercury from Various Spiked Natural Waters 27
15 Depletion of Mercury under Simulated Anaerobic Conditions 28
16 Effect of Washing Fibers on Mercury Sorption 29
vn
-------
SECTION I
CONCLUSIONS
1. Waste wool fiber was found to be an effective scavenger for mercury
in waters over a wide concentration range. More importantly, signifi-
cant amounts of mercury were taken up by wool from a sample of natural
polluted sludge. At the 1-ppm level of mercury in water used in most
of our experiments, 90-95% of organic or inorganic mercury was removed
within 24 hours. Large quantities of mercury could be removed at higher
concentrations up to 300 mg/gram of fiber. Wool fibers function both
by themselves and in wool/polyester fiber blends. Accordingly, waste
wool has potential for removing mercury from naturally contaminated
waters and bottom deposits and provides a use for industrial waste fiber.
2. Nylon fiber is not as effective as wool, removing 20-50% of the
mercury depending upon circumstances. No essential difference is noted
between the two major types of nylon—nylon 6 and nylon 66.
3. Some variability in the effectiveness of types of wool was found,
as would be expected. Wool with certain fiber finishes showed decreased
effectiveness but by washing the fiber prior to use, effectiveness could
be restored.
4. For practical purposes, wool sorbs mercury species from several
sources about equally well. Sources of mercury tested in this program
were mercuric chloride, methyl mercuric chloride, phenyl mercuric acetate,
bis (2-methoxyethyl) mercury, and dissolved metallic mercury.
5. Variations in pH (2-10) and temperature (5-35°C) do not greatly alter
sorption of mercury. Water hardness also has little effect on the effec-
tiveness of wool, and sorption occurs under simulated anaerobic condi-
tions.
6. The presence of sulfide in water or sediment does reduce the effi-
cacy of wool, probably due to the fact that it forms insoluble mercuric
sulfide and thereby limits the availability of mercury.
-------
SECTION II
RECOMMENDATIONS
On the basis of the data obtained in our laboratory studies, we recommend
two avenues for future work. Possibly both can be carried on concurrently.
The first is to fabricate selected wool fibers into suitable forms
(weighted blankets or felts, plugs, nets, etc.) to be placed in field
sites with known mercury contamination to measure mercury removal
under real conditions. Although several types of wool structures should
be tested at first, nonwoven wool felts would be the preferred substrate,
since they are inexpensive to make and have good strength and surface
area. Appropriate analytical measurements for mercury in the environment
and on the fiber would be made, along with determinations of salinity,
hardness, temperature, and other properties of the water. Consideration
should be given to disposal of the contaminated fibers or chemical recla-
mation of mercury from them.
The second concerns a more detailed study of factors affecting mercury
removal. This might emphasize the role of sulfides and possibly other
species not yet investigated. The broader use of wool to scavenge other
metallic ions such as cadmium should be included. Methods for increasing
the effectiveness of nylon should be investigated, since waste nylon is
available in great quantity. Mercury depletion at much lower levels
should be investigated.
-------
SECTION III
INTRODUCTION
Much technical knowledge exists describing the sorption of metallic ions
by proteinaceous or keratinaceous materials, chief of which are wool and
hair (Appendix A). This attraction between proteinaceous fibers and
metals is important to certain technologies such as the old fur and
wool felt hat industry, where mercury salts were used to felt wool, and
to the textile dyeing and finishing industry, where dyestuffs and
finishes containing metallic sites are affixed to fibers.
A paper, "Sorption Behavior of Mercuric and Methylmercuric Salts on Wool,"
by Friedman, et al., presented at the American Chemical Society Meeting
in Los Angeles, March 29-April 2, 1971, describes a laboratory study of
various products as sorbents for mercury ions in water. It was concluded
that wool was especially effective in removing dissolved mercury under
certain conditions and that it might be useful for decontamination.
Therefore, it seems reasonable to utilize keratinaceous or proteinaceous
materials like wool as scavengers for mercury salts in water. Besides
wool, we include nylon fiber, which can be considered a man-made protein-
aceous material. It resembles wool in some respects, due to the presence
of amide linkages in the polymer along with amine and carboxylic acid
side- or end-groups. Nylon, of course, does not contain the disulfide
or sulfhydryl groups found in wool.
The use of such fibers is made more interesting when one considers the
large quantities of waste fiber and other proteinaceous materials that
are potentially available. Although the use of wool has diminished in
recent years, large quantities of waste wool are still available. Growth
in the use of synthetic and nonwoven fibers is creating a major solid
waste problem. For instance, when synthetic fibers are used in blends
with wool, the waste product can no longer be reclaimed. In 1967, about
1.7 billion pounds of textile waste was generated. Wool/polyester blend,
a common textile blend, represents a part of this figure. Nylon
producers' waste, about 150 million pounds per year, is also reported
to be available.
The primary goal of this program was to study the feasibility of using
proteinaceous fibers from industrial waste as scavengers for mercury
introduced into streams and lakes by contaminated bottom deposits.
Although many proteinaceous materials are available, our investigation
was limited to wool, nylon, and wool/polyester blend waste.
-------
To simplify the feasibility study, our work was conducted in the labora-
tory. We are aware that actual application depends upon certain rate
processes in the aquatic environment, degree of contact between protein
and mercury, physical form of fiber, etc. Competition for mercury and
other sorbable metallic ions between protein fibers and bottom muds
certainly would be a final factor in efficacy. However, before these
problems can be considered, certain basic information must be obtained
in the laboratory. The chemistry of the mercury conversion and use
by organisms in the aquatic environment is still a subject of study
elsewhere. It has not been our intent to center on the theoretical or
the mechanistic nature of the process, but only to demonstrate the
efficacy of mercury sorption by the protein materials. We are looking
for large effects which would indicate the practical usefulness of the
fibers.
Briefly, the program consisted of obtaining representative waste fibers
from industrial sources and then measuring the removal of mercury by
these fibers from solutions containing a variety of mercury species
under a range of conditions that might be expected to influence mercury
activity. The parameters evaluated in this study included the following:
Protein Fibers - Wool, nylon, and wool/polyester blend wastes were
obtained from commercial textile sources. Because dyeing of fibers
can sometimes alter their properties, we inspected both dyed and un-
dyed fibers.
Type of Mercury Compound - For demonstrative purposes, we used mercuric
chloride, methyl mercuric chloride, phenyl mercuric acetate, bis
(2-methoxyethyl) mercury, and metallic mercury as sources of mercury
in solution. Except for the metallic mercury, these compounds are
relatively soluble in water. All five might be expected to represent
species found in the aquatic environment. Metallic mercury is less
soluble than the other compounds but does have some solubility in water.
The levels of mercury may, in practice, range from parts per billion
(ppb) to hundreds of parts per million (ppm). For most of our work,
we selected one part per million as a suitable concentration. The
fiber to liquor ratio was selected for the most part to be 1 part
fiber to 200 parts of liquid by weight.
_p_H - We would expect sorption from acid and alkaline waters to be
different; therefore, measurements were made in buffered solutions of
various pH.
-------
Temperature - Temperature might also be expected to affect sorption.
Experiments were made at temperatures of 5°, 23°, and 35°C to observe
changes in sorption.
Time - Measurements were made in many cases over a period of hours and
days in order to observe sorption behavior of mercury as a function of
time.
Quality of Water - Since a variety of other chemical species in water
can be expected to affect mercury sorption, we obtained natural waters
from known sites in Massachusetts to be spiked with mercury.
-------
SECTION IV
DESCRIPTION OF FIBERS
A selection of typical waste fibers was obtained from industrial sources
for use in this work. Their nature and source are identified in Table 1.
The three fibers from Arthur D. Little (ADL), Inc., were in-house mate-
rials which are not fully identifiable but were employed for introductory
work prior to receiving the other industrial fibers. For simplicity,
the fiber number and generic descriptions are used throughout the report.
TABLE l
FIBERS TESTED
Fiber
No. 1
No. 2
No. 3
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
No. 9A
No. 10
No. 11
No. 12
No. 13
Description
Nylon-66 (100%) fiberstock waste,
1-15 MMM, regular, waste reference
No. 82840, white
Nylon-6 (100%) reclaimable nylon
thread waste, merge 824, white
Nylon-66 soft waste from warper
section beam, 40/34/6.0 Z SD T280
Dup. Nyl., Lot No. 1270, white
Scoured virgin wool
Virgin polyester staple, 3 denier
Polyester/wool (55%/45%) reprocessed
rags, semi-open (picked), in mixed
dark colors
Mixed wool knit rags (70-80% wool),
semi-open (picked), in mixed dark
colors
100% wool comber noils, 70's A/0
100% wool spinning waste (Pneumaf11),
undyed
Like Fiber No. 9 but dirtier in
appearance
Polyester/wool (60%/40%), waste,
code No. ML 105, grade or mix
No. 6040 T-65 Dae. 40% wool 64's,
color No. tint, name or type is
Pneumafil
Virgin wool top, grade 70's
Polyester/wool blend, dyed
Virgin nylon stable, type 200
Source
E. I. du Pont de Nemours & Co., Inc.
Seaford Plant, Delaware
American Enka Corporation
Lowland, Tennessee
Vinton Weaving Company
Division of Burlington Industries
P. 0. Box 337
Vinton, Virginia
Buckley &
Franklin,
Buckley &
Franklin,
Buckley &
Franklin,
Mann, Inc.
Massachusetts
Mann, Inc.
Massachusetts
Mann, Inc.
Massachusetts
Buckley & Mann, Inc.
Franklin, Massachusetts
Burlington Worsteds
Division of Burlington Industries
Clarkesville, Virginia
Burlington Worsteds
Division of Burlington Industries
Clarkesville, Virginia
Burlington Worsteds
Division of Burlington Industries
Clarkesville, Virginia
Burlington Worsteds
Division of Burlington Industries
Clarkesville, Virginia
Arthur D. Little, Inc.
Arthur D. Little, Inc.
Arthur D. Little, Inc.
-------
In all, the fibers included virgin wool, used wool, dyed wool, undyed
wool, processed wool, polyester/wool blend, polyester, and two types
of nylon. Polyester fiber alone was not expected to sorb mercury
but was included for comparison with the widely used polyester/wool
blends.
The mercury content of selected fibers (Table 2) was measured to
assure ourselves that the fibers did not already contain levels of
mercury which might influence experimental results. The method of
measurement (Procedure 2) and other experimental details are given
in Appendix B. The highest level of mercury occurred in wool (Fiber
7) that came from dyed, processed rag materials; by their nature,
such materials have a history conducive to sorption of metallic ions
from dyebaths, scouring and finishing baths, etc. Polyester/wool
(Fiber 6), also from processed, dyed rags, had the next highest level
of mercury. Undyed new fibers of polyester/wool (Fiber 10), wool
(Fibers 9 and 9A), and nylon (Fiber 1), all exhibited negligible
concentrations.
TABLE 2
MERCURY CONTENT OF SELECTED FIBERS
Mercury Found
Fiber yg Hg/g
No. 1 (1002 nylon) <0.04, <0.04
No. 6 (45% wool) 0.36, 0.5(
No. 7 (80% wool) 1.2, 1.2
No. 9 (100% wool) <0.04 <0.04
No. 9A (100% wool) <0.04, <0.04
No. 10 (40% wool) <0.04 (one sample)
In our work, we ignored the small mercury content of the fibers, since
depletion of mercury from a solution containing 1 ppm mercury takes
place in any case. In fact, Fiber 7, which contained the highest
level of mercury, was found to be among the most active of the sca-
vengers.
10
-------
SECTION V
SORPTION OF MERCURY BY FIBERS
The various fibers were screened for their effectiveness in reducing
mercury levels in solutions prepared with inorganic and organic sources
of mercury (mercuric chloride, methyl mercuric chloride, phenyl mercuric
acetate, bis (2-methoxyethyl) mercury, and metallic mercury). Substan-
tial amounts of mercury were removed in all cases when wool was present,
and greatest removal occurred within 24 hours. The order of effective-
ness was wool>polyester/wool>nylon>polyester. Mercury was depleted from
both organic and inorganic sources.
The general procedure for determining the propensity of fibers to remove
mercury consisted of saturating the fibers in solutions containing
appropriate mercury compounds for a given time, removing an aliquot of
the liquid, and then measuring mercury content by Flameless Atomic
Absorption (FAA) spectrometry—Procedures 1, 4, and 5, Appendix B. All
fibers as received were conditioned at 72°F and 50% R.H. prior to use.
All our work was carried out in covered polypropylene containers to
minimize loss of mercury to the environment (adsorption onto container
walls or volatilization into the atmosphere). For each experiment
throughout our work, a blank without fiber was run in polypropylene.
Loss of mercury to the environment was small but did seem to increase
slightly with time. However, since we were concerned with observing
large effects with fibers, we used the blanks only for monitoring un-
expected deviations which can occur when working with trace amounts of
metal ions.
Note that in our work we decided to ascertain sorption of mercury on
the fibers by measuring its depletion from solution. We felt that
actual analysis of mercury taken up on each fiber represented a time
consuming task, since each sample would have to be thoroughly digested
to recover mercury. To assess the validity of using mercury depletion
from solution as a measure of mercury taken up by the fiber, a material
balance between mercury in solution and mercury on the fiber was con-
ducted in selected cases. A conventional bromine oxidation method
(Procedure 3) was found suitable for removing mercury from fibers. The
recovered mercury could then be analyzed by FAA and balanced against
mercury in solution. Data for selected fibers after one week of con-
tact in mercury-containing solutions is shown in Table 3. Within experi-
mental limits, the material balance looks reasonable and indicates that
mercury depletion from solution is a valid means of monitoring sorption
of mercury by fibers.
11
-------
TABLE 3
MATERIAL BALANCE
(Results after 1 week contact in pg Hg found, based on whole sample;
200 ug originally in solution)
Mercury Source
Mercuric Chloride
Fiber
No. 4 (100% wool)
No. 6 (45% wool)
No. 7 (80% wool)
Measured
Mercury
Taken up
By Fiber
160
260
210
Measured
Mercury
Removed
From
Liquid
190
190
200
Percent of
Available
Mercury
Sorbed
By Fiber
(fiber/ liquid)
80
130
100
Methyl Mercuric Acetate
No. 1 (100% nylon) 41,31,20 60
(30 average)
No. 4 (100% wool) 160,239,233 190
(210 average)
No. 6 (45% wool) 180
No. 7 (80% wool) 210
180
190
60
110
100
110
Phenyl Mercuric Acetate
No. 4 (100X wool)
No. 6 (45% wool)
No. 7 (80% wool)
170
260
188,242,196
(210 average)
190
190
190
90
UO
110
Initially, Fibers 11, 12, and 13 (wool, polyester/wool, and nylon) were
used to establish operational techniques (Procedure 4). The results
in Table 4 show that the wool rapidly removes mercury from solutions
containing a relatively high level of mercury, 50 ppm or 50 yg Hg/ml.
In fact, mercury levels at 24 hours had fallen to about the 1 ppm detec-
tion limit of the measuring technique using the air/acetylene flame.
The remainder of our work was carried out using solutions containing
1 ppm (1 yg Hg/ml), and the more sensitive Flameless Atomic Absorption
(FAA) technique was used for measurement.
12
-------
TABLE 4
DEPLETION OF MERCURY (50 ppm) FROM SOLUTIONS
CONTAINING FIBERS
Fiber
No. 11 (wool)
Soaking
Time (hr)
0
24
48
96
168
Measured Concentration of Mercury in Solution
(ug Hg/ml)*
No. 12 (polyester/ 0
wool) 24
48
96
168
No. 13 (nylon)
0
24
48
96
168
Mercuric Chloride
51
NA
NA
51
12
5.5
1.1
1.1
51
40
39
37
37
Phenyl Mercuric
Acetate
46
5
4.4
4.4
3.9
46
7.1
2.7
2.7
NA
46
32
31
27
23
NA = not analyzed
* Initial concentration 50 yg Hg/ml
Tables 5, 6, and 7 show the results for Fibers 1 to 10 with three
different sources of mercury (Procedure 5). Although all fibers were
screened for mercury uptake, they were evaluated in different groups,
indicated as experiments A, B, and C. The screening experiments were
conducted at ambient laboratory conditions. Experiments A and B were
done with unbuffered mercury solutions, while in experiment C the
solutions were buffered to pH 6.2, which is close to that found for
water saturated with CC>2 from air. Experiments A and B compared
fibers in a general way; experiment C provided a more detailed view of
mercury depletion with time.
Results from experiment C, which are considered typical, are graphed
in Figures 1 and 2. Fiber 7 (wool) removed essentially all of the
inorganic mercury almost immediately, and the organic mercury was ad-
sorbed within four hours. Fiber 9 (also wool) removed over 50% of
the inorganic mercury within the first several hours and was as effec-
tive as Fiber 7 after about 24 hours in removal of organic mercury.
Nylon fiber (Fiber 1) was much less efficient in depleting mercury.
The controls are the polypropylene containers.
13
-------
TABLE 5
Fiber
No. 1
No. 2
No. 3
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
No. 9A
No. 10
DEPLETION OF MERCURY (1
Experiment
(100% nylon 66) A
B
C
(100% nylon 6) A
B
C
(100% nylon 66) A
B
C
(100% wool) A
B
C
(100% polyester) A
B
C
(45% wool/50% polyester) A
B
C
(80% wool) A
B
C
(100% wool) A
B
C
(100% wool) A
B
C
(100% wool) A
B
C
(40% wool/60% polyester) A
B
C
pptn) FROM SOLUTION CONTAINING MERCURIC CHLORIDE
Measured Concentration of Mercury in Solution (pg Hg/ml)*
1 hr 2 hr 4 hr 6 hr 24 hr 48 hr 168 hr
0.12 0.04
0.64 0.38 0.22
0.90 0.92 0.92 0.82 0.60 0.64 0.56
0.26 0.15
0.14 0.09
0.08 0.03
0.50 0.64
0.07 0.04
0.06 0.02
0.09 0.07 0.07 0.15 0.05 0.04 0.03
0.52 0.41 0.18
0.57 0.59 0.36
0.47 0.48 0.49 0.46 0.37 0.33 0.21
0.57 0.50 0.30
0.37 0.35 0.20
Blank (polypropylene beaker) A 0.47 0.84
B 1.04 1.04 0.94
C 0.97 0.97 0.98 0.96 0.88 0.88 0.74
* Initial concentration = 1.0 pg Hg/ml
-------
TABLE 6
I-1
Ui
DEPLETION OF MERCURY (1
Fiber
No.
No.
No.
No.
No.
No.
No.
No.
No.
No.
No.
1 (100% nylon 66)
2 (100% nylon 6)
3 (100% nylon 66)
4 (100% wool)
5 (100% polyester)
6 (45% wool/55% polyester)
7 (80% wool)
8 (100% wool)
9 (100% wool)
9A (100% wool)
10 (40% wool/60% polyester)
Experime
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
ppm) FROM SOLUTION CONTAINING METHYL MERCURIC CHLORIDE
nt Measured Concentration of Mercury in Solution (yg Hg/ml)*
1 hr 2 hr 4 hr 6 hr 24 hr
0.69
1.00 0.86
0.90 0.97 0.94 0.86 0.84
0.80
0.71
0.04
0.87
0.24
0.05
0.28 0.18 0.08 0.06 0.07
0.09 0.05
0.07 0.03
0.75 0.62 0.53 0.38 0.03
0.09 0.05
0.16 0.07
48 hr 168 hr
0.72
0.90
0.82 0.94
0.69
0.64
0.04
0.88
0.09
0.04
0.06 0.03
0.02
0.02
0.01 0.02
0.02
0.08
Blank (polypropylene beaker) A 0.83 0.93
B 1.08 0.98 0.98
C 0.96 0.95 0.97 0.94 0.90 0.88 0.98
* Initial concentration = 1.0 ug Hg/ml
-------
TABLE 7
DEPLETION OF MERCURY (1
Fiber
No. 1
No. 2
No. 3
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
No. 9A
No. 10
Blank
ppm) FROM SOLUTION CONTAINING PHENYL MERCURIC ACETATE
Experiment Measured Concentration of Mercury in Solution (pg Hg/ml)*
(100% nylon 66)
(100% nylon 6)
(100% nylon 66)
(100% wool)
(100% polyester)
(45% wool/55% polyester)
(80% wool)
(100% wool)
(100% wool)
(100% wool)
(40% wool/60% polyester)
(polypropylene beaker)
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
6 hr 24 hr
0.66
0.80,0.84 0.72
0.50
0.38
0.05
0.84
0.18
0.06
0.10 0.05
0.06 0.03
0.06 0.05
0.12 0.06
0.87
1.10 1.12
48 hr 168 hr
0.48
0.58
0.32
0.20
0.05
0.82
0.05
0.03
0.04
0.02
0.03
0.03
0.98
0.90
* Initial concentration = 1.0 pg Hg/ml
-------
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
Time (hr)
FIGURE 1 DEPLETION OF MERCURY FROM MERCURIC CHLORIDE SOLUTION
-------
00
Control (Polypropylene)
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
Time (hr)
FIGURE 2 DEPLETION OF MERCURY FROM METHYL MECURIC CHLORIDE SOLUTION
-------
With Fibers 1-10, results in the tables generally indicate that wools
remove a substantial amount of mercury within 24 hours. The wool in
Fibers 9 and 9A, which is characterized by poor sorption compared with
other wools, likely contains a fiber finish and was found to be more
effective when the finish was removed (Section VI). Nylons appear to
be less effective mercury scavengers. Inspection of the data at 168
hours indicates that nylon removes inorganic mercury better than organic
mercury. The polyester fiber (Fiber 5) does not remove significant
mercury, as indicated when concentration values are compared with appro-
priate blanks. Note that the low values for polyester and blank in
Table 5 appear to be anomalous. Although mercury continues to be
depleted from solution after seven days in ail cases, the major percen-
tage of depletion occurs during the first 24 hours.
In addition to the three mercury sources already described, fibers were
screened in the presence of bis (2-methoxyethyl) mercury and soluble
metallic mercury (Procedure 5). The results are listed in Table 8.
Again, the wool remains an effective scavenger of mercury, even for solu-
bilized metallic mercury. Note the improvement of Fiber 9 when washed.
Like the other salts, the original mercury level was 1 ppm in the case
of bis (2-methoxyethyl) mercury. In the case of metallic mercury, the
water over a pool of mercury was determined to contain about 60 ppm Hg
as solubilized metal, and this solution was adjusted to contain about
1.5 ppm for the other experiment. Assuming in this case that the blank
represents 100% of available mercury in solution, we observe Fiber 9
removes 17% of available solubilized metallic mercury and Fiber 7 removes
57%. Thus, Fiber 7 not only has a high capacity for organic and in-
organic ionic mercury, but also for solubilized metallic mercury.
TABLE 8
DEPLETION OF MERCURY FROM SOLUTIONS CONTAINING
BIS(2-METHOXYETHYL) MERCURY AND DISSOLVED
METALLIC MERCURY
(Results are in pg Kg/ml Remaining in Solution After 24 Hours)
Bis (2-Methoxyethyl)* Metallic**
Fiber Mercury Mercury
No. 7 (80% wool) 0.15 0.65
No. 9 (100% wool) 0.47 1.25
No. 9 (washed - 100% wool) 0.16
Blank (Polypropylene Beaker) 0.91 1.50
* Initial concentration = 1.0 ug Hg/ml
** Initial concentration " 1.5 ug Hg/ml
19
-------
SECTION VI
FACTORS RELATING TO MERCURY SORPTION
To gain further insight into the feasibility of using the proteinaceous
fibers to scavenge mercury, we investigated in the laboratory various
factors which might affect mercury sorption. The results of these in-
vestigations are described below.
EFFECT OF pH
To study the effect of pH on the mercury uptake, we decided to use a
single buffer which could be adjusted to control pH over a wide range.
A multi-component system such as the usual "universal" buffers, which
may contain phthalate, chloride, acetate, citrate, carbonate, etc., was
not considered desirable due to the many possible interactions and com-
plex formation with the mercury species. In addition, phthalate and
acetate are not usually constituents of natural water.
A dilute phosphate solution seemed most appropriate, since it would
involve only two of the four possible phosphates (acid, mono, di, and
tri salts) at any one pH, it would have some buffer capacity over most
of the pH range, and phosphates are natural constituents of water
(albeit at low levels).
A concentration of 0.01 M total phosphate was selected, although we
recognized that it was higher than normally found in water. We selected
pH 4, 6, and 8 as the probable extremes of natural water but also added
a few points at pH 2 and 10 to look for any unexpected behavior. The
buffered solutions were prepared with the appropriate mercury salts
and mercury sorption by fibers was measured in the usual manner (Procedure
6).
Results (Table 9) for the two organo-mercury species are consistent;
depletion of mercury from solution appears to be little affected by
large variations in acidity. In the case of the inorganic mercury
species, depletion seems to be slightly reduced as pH is increased, but
effectiveness for removing mercury still remains high. As indicated
before, performance of wool Fiber 9 remains diminished unless it is
washed.
As a matter of information, the pH of unbuffered mercury solutions
used to saturate Fibers 1 to 7 for 168 hours in a previous experiment
was measured. In general, for a given fiber, pH remained about the
same for solutions of the three mercury compounds (Table 10). In this
case, differences in pH among fibers may represent the pH contribution
from the fibers themselves.
21
-------
TABLE 9
Mercury Source
Mercuric Chloride
M
Fiber
No. 1 (100% nylon)
No. 7 (80% wool)
No. 9 (100% wool)
No. 10 (40% wool)
Methyl Mercuric Chloride No. 7 (100% nylon)
No. 7 (80% wool)
No. 9 (100% wool)
No. 10 (40% wool)
Phenyl Mercuric Acetate No. 9 (100% wool)
DEPLETION OF MERCURY AT VARIOUS
ition of Mercury in
2
<0.01
0.06,0
—
'.ne Beaker) 0.93,0
—
—
0.04
—
:ne Beaker) 1.04
0.03
:ne Beaker) 0.99
Solution, yg
4
0.53
pH
Hg/ml at 24
pH
6
0.60
0.05,0.01** 0.07,0.05
.02 0.48,0.35
0.05,0.07
.98 0.93,0.96
0.94
0.08
0.04
0.10
0.97
0.03
0.99
0.45,0.44
— ,0.18
0.88,1.01
0.94
0.06
0.06
—
1.01
0.03
0.91
Hours)*
8 10
0.81
0.20,0.15 — ,0.14
0.70,0.57 0.63,0.61
0.40,0.40 —
0.88,1.00 0.98,1.05
0.94
0.07
0.04 0.04
0.09
0.96 1.01
0.02 0.02
0.97 0.97
Unbuffered
—
0.08
0.57
—
0.97
*
**
Initial concentration
Replicates
1.0 ug Hg/ml
-------
TABLE 10
pH OF UNBUFFERED SOLUTIONS CONTAINING FIBERS
(at 168 Hours)
Mercury Compound
Fiber
No.
No.
No.
No.
No.
No.
No.
1
2
3
4
5
6
7
(100%
(100%
(100%
(100%
(100%
nylon)
nylon)
nylon)
wool)
polyester)
(45% wool)
(80% wool)
Blank (Poly]
propylene
Mercuric
Chloride
6
5
5
4
4
5
4
4
.3
.9
.4
.8
.9
.6
.7
.9
Phenyl Mercuric Methyl Mercuric
Acetate Chloride
5
6
5
4
4
5
4
4
.0
.3
.6
.7
.5
.5
.6
.9
4
5
5
4
4
5
4
4
.8
.9
.3
.8
.5
.4
.9
.5
Beaker)
EFFECT OF TEMPERATURE
Fibers 1, 7, and 9 were selected for a study of the effect of tempera-
ture on mercury uptake from both mercuric chloride and methyl mercuric
chloride solutions. Fibers were saturated at 5°, 23° (room temperature),
and 35°C, and data were obtained at 24 and 48 hours in solutions buffered
to pH 6.2 (Procedure 7).
The results (Table 11) indicate that mercury depletion increases with
increasing temperature within the time period of the experiment.
However, a significant temperature effect is not observed for Fiber 7,
which is an active scavenger of mercury. Fiber 9, which is shown to
have a retarding fiber finish, shows the effect of temperature more
markedly. In the case of Fiber 1 (nylon), increasing temperature had a
greater effect in depleting the inorganic mercury matter than the organic
form. For practical purposes, wool scavenges mercury in both cold and
warm water effectively.
23
-------
TABLE 11
Mercury Source
Mercuric Chloride
Methyl Mercuric Chloride
DEPLETION OF MERCURY AT
SEVERAL
TEMPERATURES
(Results are in ug Hg/ml Remaining in Solution)*
Time
Fiber (hr)
No. 1
No. 7
No. 9
Blank
No. 1
No. 7
No. 9
Blank
(100% nylon)
(80% wool)
(100% wool)
(Polypropylene
Beaker)
(100% nylon)
(80% wool)
(100% wool)
(Polypropylene
Beaker)
24
48
24
48
24
48
24
48
24
48
24
48
24
48
24
48
0.
0.
0.
0.
1.
0.
1.
0.
0.
0.
1.
0.
b°C
76
60
0.08
0.09
44
20
02 1.00
88 0.90
06
67
0.07
0.06
22
07
10 0.98
76 0.98
Temperature
0.
0.
0.
0.
1.
0.
1.
0.
0.
0.
1.
0.
2'3"C
70
54
0.07
0.04
39
12
14
94
06
67
0.06
0.03
04
02
14
77
0
0
0
0
1
0
0
0
0
0
1
0
.32
.48
.12
.07
.00
.90
.99
.65
.07
.05
.00
.78
5"C
0.07
0.04
0.86
0.84
0.05
0.07
0.99
0.98
* Initial concentration - 1.0 ug Hg/ml
-------
CAPACITY FOR MERCURY
Two types of experiments were conducted on selected fibers (7, 9, and
10) to examine fiber capacity for mercury. We first measured the total
depletion of mercury by fibers as mercury concentration was increased,
and then we measured the continued capacity of the fibers to remove
mercury repeatedly from a series of 1-ppm solutions.
In the first experiment, one gram of selected fibers was placed in
200 ml of buffered solutions (pH 6.2) containing 5000, 1000, 100 or
10 yg Hg/ml, and mercury depletion from solution was measured in the
usual manner at 5, 24, 48 and 96 hours (Procedure 8).
When the concentration of mercury in solution was increased (Table 12),
the percent uptake of mercury by a given quantity of wool was reduced,
although larger amounts of mercury were sorbed. For a given amount of
fiber, effectiveness was greater at lower concentration, but the addi-
tion of more fiber might be expected to reduce mercury further. Close
inspection of the data indicates in several instances that the amount
of mercury in solution increases with time. There is not enough data
to determine if this is an anomaly or if sorbed mercury indeed is re-
generated from wool. Previous data with 1 yg Hg/ml solutions have al-
ways indicated a general loss of mercury with time. As a benchmark,
calculations show that one gram of fiber appears to adsorb about
300 mg Hg after 24 hours' soaking in a solution originally containing
one gram of mercury. This amounts to about 30% of the mercury. At
lower concentrations, about 95% of the mercury is removed from solution.
TABLE 12
DEPLETION OF MERCURY AT VARIOUS CONCENTRATIONS OF MERCURIC CHLORIDE
(Results are in ug/ml remaining in solution;
figures in parenthesis are percent of initial)
Fiber
No. 7 (80% wool)
No. 9 (100% wool)
No. 10 (40% wool)
Initial
Concentration
(ug/ml)
5000
1000
100
10
5000
1000
100
10
5000
1000
100
10
Concentrat Ion at Various Times
5 hr
3640(73)
630(63)
14(14)
1.1(11)
3740(75)
515(52)
5(5)
0.4(4)
3920(78)
525(53)
25(25)
0.3(3)
24 hr
3840(77)
575(58)
4(4)
0.2(2)
3500(70)
455(46)
2(2)
0.4(4)
2720(55)
595(60)
5(5)
0.6(6)
48 hr
3560(71)
545(55)
2(2)
1.5(15)
3300(66)
390(39)
0.9(9)
3600(72)
580(58)
2(2)
0.8(8)
96 lir
3500(70)
465(47)
2.7(27)
3360(68)
430(43)
2.3(23)
3800(76)
630(63)
0.5(.5)
0.5(5)
Blank (Polypropylene Beaker) 10
9.65
9.65
25
-------
In the second experiment, the fiber samples were saturated five
different times, 24 hours apart, in fresh solutions of mercuric
chloride containing 1 yg Hg/ml and buffered to pH 6.2 (Procedure 8).
Depletion of mercury from solution was measured in the usual manner.
When the wool samples were saturated in 1 yg Hg/ml solutions five
consecutive times (Table 13), they removed about 95% of the mercury
each time. In other words, the wools had taken up about 1 mg of mer-
cury per gram of fiber with no apparent loss of effectiveness for
mercury removal at the 1 ppm level.
TABLE 13
DEPLETION OF MERCURY BY FIBERS
WITH REPEATED CONTACT WITH 1 ug/ml MERCURIC CHLORIDE SOLUTION
(Results are In ug/ml Remaining in Solution)
Fiber Number of Contacts
1 1 1 i 1
No. 7 (80% wool) 0.07 0.06 0.08 0.11 0.10
No. 9 (100% wool) 0.37 0,09 0.04 0.04 0.03
No. 10 (40% wool) 0.13 0.05 0.05 0.05 0.05
Blank (Polypropylene Beaker) 0.85 0.88 0.95 0.96 0.93
Although mercury is depleted from solutions with high mercury content,
it appears that the efficiency of a fiber increases as the level of
mercury decreases.
QUALITY OF WATER
In addition to the other factors investigated, such as temperature
and PH, we believed that the quality of water might have an effect
upon mercury depletion. For this purpose, we collected four natural
waters of varying quality to be artificially contaminated with mer-
cury. All were analyzed for pH and hardness, with the following results:
Hardness
pH (mg/1
Horn Pond, Woburn 7.7 95
Babson Reservoir, 6.7 36
Gloucester
From a woodland pond, 5.9 IQ
Sudbury
Atlantic Ocean, off Essex 7.8 1270
26
-------
The Sudbury water exhibited a yellow color and an obvious sulfidy odor.
The other waters had no significant odor or color.
The waters were used as received. They were spiked with mercuric chloride
(1 ug Hg/ml) prior to addition of fibers. Measurements at 24 hours
were made in the usual manner (Procedure 9).
From the data in Table 14, we can see that water hardness per se and
pH had little effect upon mercury depletion, since the wool fibers
scavenged to a similar extent in the natural waters and in distilled
water, except in the case of the Sudbury water, which was obviously
high in sulfur. The sulfide character of this water may have decreased
availability of mercury due to formation of a stable complex or it may
have decreased the ability of the fibers to adsorb mercury due to inter-
action with fibers. The decreased blank in this case could indicate
formation of insoluble HgS.
TABLE 14
DEPLETION OF MERCURY FROM VARIOUS SPIKED NATURAL WATERS
(Measured Concentration of Mercury in Mercuric Chloride Solution, yg Hg/ml)*
Source
Fiber
No. 7 (80% wool)
No. 9 (100% wool)
No. 9 (washed - 100% wool)
Blank (Polypropylene Beaker)
* Initial concentration - 1.0 yg Hg/ml
** For comparison, data taken from Tables 5 and 16.
ANAEROBIC CONDITION
We ran a brief experiment to see if mercury depletion under simulated
anaerobic conditions differed from the depletion commonly observed from
solutions saturated with air. Mercuric chloride and methyl mercuric
chloride were the choice as model compounds, and Fibers 7, 9, and 9
(washed) were used. Solutions were buffered to pH 6.2 using 0.01 M
phosphate. Nitrogen was bubbled through the solutions to displace air
and a blanket of nitrogen was maintained over the solutions throughout
the experiment. At 24 hours, mercury depletion in solution was measured
in the usual manner (Procedure 10).
Depletion data under anaerobic conditions are shown in Table 15 and
comparison is made with typical aerobic data. We found no significant
difference from data obtained under air-saturated conditions.
27
Woburn
0.08
0.45
0.16
0.83
Gloucester
0.03
0.45
0.05
0.95
Sudbury
0.48
0.61
0.45
0.68
Atlantic
Ocean
0.06
0.30
0.12
0.93
Distilled*
Water
0.05
0.37
0.10
-0.90
-------
TABLE 15
DEPLETION OF MERCURY UNDER SIMULATED ANAEROBIC CONDITIONS
(Measured Concentration of Mercury In Solution, yg Kg/ml at 24 Hours)*
Solution
Fiber
No. 7 (80% wool)
No. 9 (100% wool)
No. 9 (washed - 100% wool)
Blank (Polypropylene Beaker)
Mercuric Chloride
Methyl Mercuric Chloride
Anaerobic Aerobic** Anaerobic Aerobic**
0.16
0.45
0.12
1.07
0.05
0.37
0.10
-0.90
0.05
0.05
0.05
1.10
0.07
0.03
—
~0.90
* Initial concentration 1.0 yg Hg/ml
** Aerobic data taken from Tables 5, 6, 16
EFFECT OF WASHING FIBER
Data in previous sections indicated that Fiber 9 (100% wool) did not
sorb mercury from mercuric chloride solution as effectively as other
wool fibers, although sorption from methyl mercuric chloride seemed
normal. This behavior was thought to be due to the presence of a
fiber finish which might be hydrophobic in nature. Fiber finishes are
normally composed of oils or modified oils and might retard sorption
of inorganic materials. To resolve this point, we scoured Fiber 9 and
several others with a typical wool cleaning solution composed of deter-
gents to observe how washing of fibers would affect subsequent mercury
sorption (Procedure 11). Depletion of mercury from solution was
measured in the presence of the washed fibers in the usual manner using
mercury solutions containing 1 yg Hg/ml and buffered to pH 6.2.
The data in Table 16 show that washing greatly enhanced the effective-
ness of Fiber 9 (as well as that of Fibers 1, 8 and 10); the others were
relatively unchanged. Although the wash waters were not chemically
analyzed to detect a finish, it seems likely that finishes on these
fibers are the cause of the noted behavior.
28
-------
TABLE 16
EFFECT OF WASHING FIBERS ON MERCURY SORPTION
(Results Are in pg Hg/ml Remaining in Solution After 24 Hours)*
Fiber
No. 1 (100% nylon)
No. 4 (100% wool)
No. 7 (80% dyed wool)
No. 8 (100% wool)
No. 9 (100% wool)
No. 10 (40% wool)
Blank (Polypropylene Beaker)
Mercury Source
Mercuric
Washed
0.32
0.07
0.03
0.13
0.10
0.05
—
Chloride
Unwashed**
0.60
0.08
0.05
0.41
0.37
0.37
0.88
Methyl
Washed
1.0
0.07
0.06
0.06
—
0.06
1.10
Mercuric Chloride
Unwashed**
0.84
0.04
0.07
0.05
—
0.07
0.90
* Initial concentration = 1.0 yg/ml
** Values from Table 4
29
-------
SECTION VII
MERCURY REMOVAL FROM SEDIMENT
To establish that mercury could be removed by wool from a naturally
contaminated sediment, we obtained a sample of bottom sludge taken by
JBF Scientific Corporation from a site in the Ashland Reservoir near
the Sudbury River. We realize that bottom sludges are biologically and
chemically active and that when they are removed from the environment,
changes can occur which can affect the mercury species. However, this
brief experiment with selected wool fibers (Nos. 7, 9, and 9 washed)
indicates the potential usefulness of wool for depleting mercury from
bottom deposits even when the mercury is released slowly from the de-
posit. The rate of removal of mercury from deposits will depend on the
solubility of the mercury species.
Mercury measurements were made on solutions in the usual manner and
mercury content of sludge and fibers was determined through means of
the bromine oxidation step (Procedure 12).
Analyses of the sludge indicated about 48 yg Hg/g and a content of
0.33% sulfur (both on a wet basis). When the sludge was placed in
buffered water (pH 6.2) for 48 hours—in one case with stirring and
the other case without—and the mercury content of the water was deter-
mined, the following values were found:
Stirred 0.02, 0.03, 0.01 yg Hg/ml
Unstirred 0.01, <0.01 yg Hg/ml
Obviously, very little mercury was dissolved.
To determine if wool would remove mercury from this sludge, Fiber 7 was
mixed with sludge and water and was allowed to remain eight days. The
fiber was removed and washed. Mercury was determined in the fiber and
contact solution. The following results were obtained.
Mercury on fibers 3.20, 3.30, 3.88, 2.84 yg Hg/g
after immersion (Average = 3.31)
Mercury remaining 0.02,
-------
From the above data, it appears that wool fiber (No. 7) removes 2.3 yg
of mercury from the sludge, or 4.8% of the mercury present, in eight
days (approximately 0.5% per day). Clearly, transfer of mercury from
this sludge to fiber is slow, but nevertheless significant amounts
were removed from the sludge and were taken up by the wool.
32
-------
SECTION VIII
DISCUSSION
MECHANISM OF SORPTION
Two types of mercury takeup can be expected to occur on fibers—that
which is truly bonded to the proteinaceous molecule and that which
accompanies the takeup of water as the fibers become hydrated. Molecular
bonding of mercury is unique for wool and similar fibers and is the
basis for the proposed use of wool to scavenge mercury from contaminated
waters. Mercury deposited on the fiber from water of hydration is con-
sidered a bonus.
The mechanism for sorption of mercury ions by wool and similar mate-
rials is a subject for debate. Sorption may be associated with the
basic sites, acidic sites, amino sites, sulfur-bearing sites or any com-
bination of these. However, the importance of sulfur in the natural
fiber emerges. The high-sulfur fraction of wool, S-carboxymethylkeratin
B, is reported to be the matrix or structurally disorganized substance
surrounding the microfibrils. It is reported that during treatment of
wool with water at 60-100°C, chemical changes occur; cystyl residues
(disulfide) are converted to lanthionyl residues (a monosulfide), and
the lost sulfur becomes mobile. The sulfur content of wool is 3-4%,
and its nitrogen content is 16-17%. Wool was found to scavenge mercury
effectively. On the other hand, nylon (a synthetic polyamide having
about 16% nitrogen but no sulfur) adsorbs much less mercury, as shown
in our work. When sludge or water with a high sulfide content was used,
we observed decreased mercury sorption even by wool, indicating that
competition exists for the mercury between sulfur in wool and sulfide
or other ligands in water.
The contrast in effectiveness between nylon and wool might be further
explained by differences in the availability of the fiber molecules to
sorb mercury. Wool, a hydroscopic fiber, picks up 10-17% moisture
from air at 70°F and 65% R.H. and up to 30% from saturated air. The
nylons, on the other hand, pick up only 4-5% moisture at 70°F and 65%
R.H. Being less water-sensitive, nylon might be expected to sorb less
dissolved mercury because ion mobility is restricted in the essentially
hydrophobic nylon fiber. Polyester fiber, which has neither nitrogen
nor sulfur and picks up negligible moisture (0.4-0.8% at 70°F and 65%
R.H.), removes essentially no mercury from solution.
Although we conclude that for practical purposes wool is an effective
agent for depleting mercury from waters, the phenomenon has not been
fully explored. The effect of the presence of sulfides in waters
should be clarified, as well as the role of fiber hydrophobicity.
33
LIBRARY U.S. EPA
-------
FIELD UTILIZATION AMD COST
Because our study was conducted in the laboratory, no actual field
information is available; nevertheless, we can speculate on means of
utilization as well as costs.
Obviously, the fibers require placement into the water close to or on
the bottom, since the mechanism for mercury exchange is through solu-
tion. Although field use of wool for scavenging mercury might include
application in the form of woven cloth structures or bagged fibers,
the most practical and economical approach would seem to be through
use of nonwoven felt fabrics. Wool felt, one of the oldest forms of
fabric, is defined as "a fibrous material built up of interlocked
fibers by mechanical and chemical action, moisture, and heat. The
blend may consist of wool and other fibers" (The Felt Association,
Inc.). Felts can be made in a variety of sizes and with a specific
thickness, resiliency, firmness, etc.; they have good strength, water
permeability, and fiber surface area; and they are the least expen-
sive cloth structure to produce. If necessary, added strength can
be obtained by incorporating nylon fiber or scrim into the wool felt.
The maximum area of felt that can be introduced into the water from
a vessel and the best means of conducting the placement will have to
be investigated. Conceivably, the felt could be laid from a roll in
the boat. The felt would probably require weighting, since the
specific gravity of felts is usually less than one. In addition, to
insure vertical sinking of the felt, an open pattern or holes may be
required in the fabric, which means that a cross-lay of two felts
would be required to get full coverage.
After the felts have become saturated an appropriate length of time,
they can be withdrawn from the water. The sorbed mercury can then
be recovered from the fibers if this seems practical. Bromination
or reduction techniques, as well as treatment with citric acid and
certain chelating agents, will remove mercury. The efficiency of
mercury removal is a subject for future study.
Although it is premature to calculate the cost of laying felt in
water and later recovering it, we can estimate the finished cost of
felt on the basis of the raw materials and unit production operations.
Wool waste is available at 2-12c/lb, depending upon grade and source;
for estimation purposes, we assume that it costs 5c/lb. The unit
operations cost for production of a nonwoven felt is 20-30c/lb. (We
shall assume 25c/lb.) In the most optimistic case, therefore, we can
assume the average raw material and felt costs to total 30c/lb for a
finished felt fabric. For comparison, a woven fabric might cost $1.30
to $1.55/lb to produce.
34
-------
A 1/4-inch thick felt has a density of 2 to 4 Ib/sq yd, depending on
firmness. Therefore, to cover one acre (4840 sq yd), we would need
9680 to 19,360 pounds of felt. At 30<:/lb, the estimated cost of 1/4-
inch felt to cover one acre would be $2900 to $5800. If a 1/8-inch
felt (1 to 2 Ib/sq yd) can be used, the cost would be $1500 to $2900;
a 1/2-inch felt (3 to 8 Ib/sq yd) would cost $4400 to $12,000.
We have found in our work that 1 gram of wool fiber can sorb as much as
300 mg Hg in 24 hours, or 0.3 Ib Hg per pound of fiber. If we consider
a 1/4-inch felt covering one acre, we have an average quantity of
15,000 pounds of wool present, which means theoretically that 4500 pounds
of mercury can be sorbed. Even if the wool is 10% efficient, 450 pounds
of mercury could be removed from the environment. Based on the current
cost of mercury at $2.40/lb, the potential recovery value, excluding
reclamation costs, may be as high as $11,000 per acre, but more impor-
tantly, large quantities of mercury can be removed from the aquatic
environment.
35
-------
SECTION IX
ACKNOWLEDGEMENTS
The support of the following ADL staff members for their advisory help
in this program is gratefully acknowledged: John Funkhouser,
James Oberholtzer, Derek Till, and James Valentine.
Sincere thanks are given to Roberta Kent and Clifford Summers for con-
ducting the major portion of the chemical and analytical work, and to
Elizabeth Morton for her diligence in typing of reports.
Acknowledgement and appreciation are given to the following firms for
supplying waste fibers used in this work: American Enka Corporation,
Enka, North Carolina; Buckley & Mann, Inc., Franklin, Massachusetts;
Burlington Industries (Vinton Weaving Co. and Burlington Worsteds),
Greensboro, North Carolina; and E. I. du Pont de Nemours & Company,
Inc., Wilmington, Delaware. We also wish to thank the JBF Scientific
Corporation, Burlington, Massachusetts, for supplying a sample of
natural mercury-contaminated bottom deposit.
Finally, the support of this project by the Office of Research and
Monitoring, Environmental Protection Agency and the help provided by
Dr. Curtis C. Harlin, the Project Monitor, is gratefully acknowledged.
37
-------
SECTION X
APPENDICES
Page
A Literature Search 41
B Experimental Procedures 45
39
-------
APPENDIX A
LITERATURE SEARCH
At the start of this program, a brief literature search was conducted
in recent references and abstracts for new information which might be
useful in our work. Sources searched were:
Chemical Abstracts, 1966 to present, a major source;
Applied Science and Technology Index, 1965 to present;
Pollution Abstracts, 1970 to 1971;
Biological Abstracts, 1971;
Nuclear Science Abstracts, 1971; and
Isolated sources such as Chemical Reviews, Chemistry
and Industry, Textile Research Journal, and various
other periodicals.
The primary reference relating to the use of proteinaceous materials
such as wool to recover mercury from waters appears to be that by
Friedman, et al. (Source 1 below). The binding of mercury and other
heavy-metal salts with various proteins and chemical compounds containing
amino groups and sulfhydryl and disulfide moieties is often noted in
biochemical papers. Methods of alleviating mercury poisoning in the body
often depend on the introduction of such chemicals to combine with
mercury. Except to point to the use of proteinaceous fibers for sca-
venging mercury by analogy, most uncovered references have limited value
at this time, and many are purely theoretical studies. Ten are abstracted
below to indicate the nature of current literature and are not intended
to be inclusive.
1) M. Friedman, C. S. Harrison, W. H. Ward, and H. P. Lundgren, "Sorption
Behavior of Mercuric and Methylmercurie Salts on Wool," U. S. Depart-
ment of Agriculture, Albany, California, 1971. Paper presented
before the Division of Water, Air, and Waste Chemistry, A.C.S.,
Los Angeles, California, March 29-April 2, 1971.
Describes sorption of mercury by waste wool and other agricultural
products to remove mercury from solution. Up to 80% mercury can
be removed from solutions of mercuric chloride and methyl mercuric
chloride under certain conditions.
41
-------
2) "Resin to Purge Mercury from Body," Industrial Research, August,
1971.
Describes use of a synthetic resin containing sulfhydryl groups
to remove mercury from mice.
3) Fumiaki Kai, "The Reaction between Mercury (II) and Organic
Compounds," Bull Chem Soc Jap. _40_, No. 10 (1967), pp. 2297-2302.
Describes reactivities between HgCl2 and basic amino acids and
suggests application as means of separating amino acids.
4) S. N. Hemrajani and C. S. Narwani, "Polarographic Study of Metal
Ion Complexes with Keratin Fibers (Wool) at pH 4-5 and 30°," J_
Indian Chem Soc, 44, No. 8 (1967), pp. 704-9.
HgCl2 was thought to combine in molecular form with wool and
other salts in ionic form. Chemically modified wool combined
with different amounts of a metal. Reduced wool was most effec-
tive.
5) A. Weinstock, P. C. King, and R. E. Wutbier, "The Ion-Binding
Characteristics of Reconstituted Collagen," Biochem J, 102 (1967),
pp. 983-8.
Ion binding qualities of bivalent cations and others from salt
solutions were studied, and their relationship to properties of
reconstituted collagen was examined.
6) M. Adam, P. Fietzek, Z. Deyl, J. Rosmus, and K. Kuehn, "Investiga-
tion on the Reaction of Metals with Collagen in Vivo," Eur J
Biochem 3, No. 4 (1968), pp. 415-18.
Mercury reacts in vivo with insoluble collagen to alter its proper-
ties.
7) K. C. Tewari, J. Lee, and N. C. Li, "Zinc and Mercury (II) Inter-
action with Cytidine and Glycylglycine," Trans Faraday Soc, 66
(Pt 8) (1970), pp. 2059-76.
Complexes mercury (Il)-cytidine and mercury (Il)-glycyl-glycine
are formed.
8) M. Matsuda and B. Takeuchi, "The Interaction of Hg-H- with Deoxy-
ribonucleic Acid," J Biochem, 6l_, No. 4 (1967), pp. 523-6.
Describes methods of association of DNA with Hg-H-.
9) D. F. S. Natusch and L. J. Porter, "Direct Detection of Mercury
(II)-Thio-Ether Bonding in Complexes of Methionine -1- S-Methyl-
cysteine BH Nuclear Magnetic Resonance," J Chem Soc D (10) (1970),
pp. 596-7.
Nature of metal-ligand bonds studied showing formation of N Hg+ 0
and S Hg++ N bonds.
42
-------
10) D. M. Crothers, "Kinetics of Complex Formation of Nucleic Acids,"
Stud Biophys 24-25. 79-81 (1970).
Discusses kinetics of Hg (II) binding to nucleic acids.
43
-------
APPENDIX B
EXPERIMENTAL PROCEDURES
General techniques for conducting the experiments are described in the
procedures outlined below. Distilled water was used for preparing
solutions, except where otherwise indicated. Mercury salts were standard
reagent-grade materials obtained from the following sources:
Mercuric Chloride—obtained from Fisher Scientific Co., Pittsburgh,
Pa; certified ACS grade;
Methyl Mercuric Chloride—obtained from Alfa Inorganics, Beverly,
Mass., assay 95%;
Phenyl Mercuric Acetate—obtained from Eastman Organic Chemicals
Dept., Eastman Kodak Co., Rochester, N.Y.; and
Bis (2-methoxyethyl) Mercury—ADL laboratory compound pre-
pared from reagent-grade mercuric acetate and ethylene gas.
Ref: The Chemistry of Organometallic Compounds, by E. G.
Rochow, D. T. Hurd, and R. N. Lewis; New York: Wiley (1957),
pp. 109 ff.
PROCEDURE 1—FLAMELESS ATOMIC ABSORPTION (FAA) SPECTRQMETRY OF SOLUTIONS
The mercury in samples was determined essentially according to the pro-
cedure of Hatch and Ott (Anal Chem 40, 2085 [1968]). Two differences
are the use of bromine water in place of permanganate solution and the
use of hydroxylamine hydrochloride in 1 N t^SO^ rather than the use of
hydroxylamine sulfate directly. These changes were made due to the
nature of the samples and to attain a low background signal.
• Reagents
a) Stannous sulfate, 10% w/v in 0.5 N t^SO^
b) Sodium chloride, 30% w/v in water
c) Hydroxylamine hydrochloride, 25% w/v in 1 N
d) Sodium chloride-hydroxylamine sulfate solution—
add 60 ml of the acidic hydroxylamine solution
and 50 ml of 30% sodium chloride solution to a
500-ml volumetric flask and dilute to mark with
distilled water.
• Instrumentation— Perkin-Elmer Atomic Absorption Spectrophoto-
metric Model 303 equipped with a 6" x 1" glass cell with epoxy-sealed
quartz windows.
45
-------
• Mercury Analysis
a) A known volume of sample was pipetted into a bubbler,
which is a 100-ml round-bottom flask containing a
3/4" x 4" well in the bottom into which a tube with
a ground glass tip is inserted. (Entire apparatus
is called a Smog bubbler, Ace Glass Co.)
b) 1 ml bromine water was added and the solutions were
mixed and allowed to stand 5 minutes.
c) 20 ml of NH20H-NaCl solution was added and the solu-
tions were mixed and allowed to stand 15 minutes.
d) 10 ml stannous sulfate solution was added, and the
bubbler was immediately affixed to the tubing leading
to the quartz cell in the AA unit.
e) With the recorder running, the pump to cycle the
mercury vapor was turned on.
f) After recorder trace had peaked (usually within 1 to
2 minutes), the inlet line to the bubbler was removed
and transferred to the vent line leading to the hood.
g) When the tracing returned to the baseline (indicating
removal of all mercury vapor from the cell and connec-
ting tubing), sampling was complete.
To determine the mercury content of fibers, sludge, and other materials
containing mercury in unknown states (organic and inorganic), we used
oxidation with bromine water prior to FAA analysis as described in
the procedures.
PROCEDURE 2—MERCURY CONTENT OF FIBERS
Two 1 g portions of fiber, taken from different places in its bulk,
were treated with 50 ml of fresh bromine water for one hour. The
solutions were decanted, all the fibers were washed with about 40 ml
of water, and the solution and wash were combined and diluted to
100 ml with water. Aliquots (25 ml) were analyzed for mercury via
FAA.
PROCEDURE 3—MERCURY BALANCE FROM FIBERS
Fibers were removed from beakers containing the mercury solution and
blotted dry between pieces of Whatman #1 filter paper. Weighed por-
tions of the semi-dry fiber were reacted with bromine water in 150-
ml beakers for approximately one hour. The solutions were decanted,
the fibers were rinsed, and the total solution was made up to a known
volume. Aliquots were removed for analysis by FAA.
46
-------
PROCEDURE 4—INITIAL SCREENING OF FIBERS
Fiber (1.0 g) was saturated in 200 ml of an aqueous solution containing
50 yg Hg/ml of an appropriate mercury compound. The experiments were
conducted in covered polypropylene beakers to minimize loss of mercury
to its environment. A control without fiber was run in a polypropylene
beaker for each experiment. At appropriate time intervals, 10-ml sam-
ples of solution were withdrawn and mixed with 1 ml of concentrated
HC1. The acidic solutions were stored in capped polyethylene vials
prior to mercury analysis by atomic absorption spectrometry using air/
acetylene. As the air/acetylene method was replaced in the remainder
of the work by the more sensitive flameless method, the former is not
described in detail.
PROCEDURE 5—SCREENING OF FIBERS WITH DIFFERENT MERCURY SPECIES
The same method was used as in Procedure 4 with 1.00 ± 0.01 gram
samples of fiber, except that the solutions contained 1 yg Hg/ml and
Flameless Atomic Absorption spectrometry was used to measure mercury.
In experiments A and B, the solutions were not buffered. Experiment C
was prepared with a 6.2 pH phosphate buffer. Solutions were prepared
with appropriate quantities of mercuric chloride, methyl mercuric
chloride, and phenyl mercuric acetate. Solutions containing 1 yg/ml
of bis (2-methoxyethyl) mercury were prepared by mixing 10 ml of 20-ppm
Hg solution with 190 ml of water buffered at pH 6.2 with 0.01 M phos-
phate and adding 1.00 ± 0.01 g of fibers for a 24-hour soaking period.
Samples were then removed for FAA analysis at the appropriate time
intervals.
Some water which had been covering a pool of mercury for several weeks
was determined by FAA measurement without addition of a reducing agent
to contain approximately 60 ppm Hg as dissolved metal. Five milliliter
aliquots of this water were mixed with 195 ml of water buffered at pH
6.2 using 0.01 M phosphate, and 1.00 ± 0.01 g samples of Fibers 7 and 9
were added. The beakers were swirled, and the fibers were allowed to
soak in the solution for 24 hours. The samples were removed after 24
hours for analysis by FAA.
PROCEDURE 6—pH EXPERIMENT
Dilute phosphoric acid (0.01 M) was titrated to values of pH close to
2, 4, 6, 8, and 10 using 1 N sodium hydroxide. The resulting solution
was diluted tenfold, the actual pH was measured, and 180 ml of the
solution was added to polypropylene beakers containing 1.00 + 0.01 g of
the selected fibers. Mercury salt solution (20 ml of 10-ppm solution)
was added by pipet, and the beaker was swirled to obtain mixing of
solutions. Samples of solution (10 ml) were removed after 24 hours of
contact with fiber, and the mercury remaining in solution was determined
47
-------
by FAA. Beakers containing only solutions of salts at the various
pH's served as controls. A pH meter was used to make the measurements.
PROCEDURE 7—TEMPERATURE EFFECT
Two hundred milliliters of buffered (pH 6.2, 0.01 M phosphate) solution
containing 1.0 yg Hg/ml was mixed with 1.00 ± 0.01 g of fiber in a
400-ml polypropylene beaker. Beakers were kept at 5°C, room tempera-
ture (approximately 23°C), and 35°C prior to adding fiber. Samples
were removed after 24 and 48 hours at the three temperatures, and the
remaining solutions were analyzed for mercury by FAA.
PROCEDURE 8—MERCURY CAPACITIES
• Total Capacity—Each fiber sample (1.00 ± 0.01 g) was placed in a
400 ml polypropylene beaker, and 200 ml of solution containing either
5000, 1000, 100, or 10 yg Hg per ml of buffered solution was added.
The beakers were swirled, covered with watchglasses, and allowed to
stand. The 5000 and 1000 yg/ml Hg solutions were prepared using
weighed amounts of mercuric chloride dissolved in pH 6.2, 0.01 M phos-
phate buffer. The 100- and 10-yg concentrations were prepared by
appropriate dilutions of the concentrates. Samples were taken after
5, 24, 48, and 96 hours' soaking time. The beakers were swirled to
ensure solution homogeneity just before sampling, and 0.1 ml of con-
centrated HC1 was added to each sample removed to prevent loss of
mercury to the container. Mercury measurement was by FAA.
• Repeat Contact Capacity—A 200-ml quantity of solution containing
1 yg/ml of mercuric chloride was buffered to pH 6.2 with 0.01 M phos-
phate. Fiber (1.00 ± 0.01 g) was soaked in this solution for 24 hours.
The beaker was swirled, and a sample was removed for analysis. The
remainder of the solution was decanted, the fiber was squeezed to
remove excess liquid, and a fresh 200 ml of 1 yg Hg/ml, pH 6.2, solu-
tion was added to the now one-day old fiber. This process was re-
peated for each fiber over a five-day period. The mercury content
of the samples was determined by FAA.
PROCEDURE 9—WATER QUALITY
Each water sample (180 ml) was mixed with 20 ml of a solution contain-
ing 10 yg Hg/ml (as HgCl2) to give a concentration of 1 yg Hg/ml.
Fiber (1.00 ± 0.01 g) was added, and the beakers were swirled and
allowed to stand 24 hours. The beakers were again swirled just before
the solutions were subjected to FAA analysis for mercury.
48
-------
PROCEDURE 10—ANAEROBIC CONDITION BUFFERED TO pH 6.2
Nitrogen was bubbled through 200 ml of solution containing 1 yg Hg/ml
to displace oxygen. Fiber (1.00 ± .01 g) was added and purging was
continued for 3 to 4 minutes. During the 24-hour saturation period, a
blanket of nitrogen was maintained over the beakers in a glove bag
filled with nitrogen. Aliquots of liquid were then removed, and the
mercury was measured by FAA.
PROCEDURE 11—FIBER WASH
Wash solution—Trisodium phosphate 0.3 gram
Triton X-100 0.1 gram
Tap water (110°F) 1000 ml
Fibers (15 g) were placed in the hot solution and periodically stirred
over a 30-minute period. They were then removed, squeezed, thoroughly
rinsed in warm tap water, and dried at 72°F, 50% R.H. The fibers were
tested in the usual way in solutions containing 1 pg/ml of mercury as
mercuric chloride and methyl mercuric chloride. Both solutions were
buffered with 0.01 M phosphate to pH 6.2. Fiber (1.00 ± 0.01 g) was
soaked in 200 ml of the above solutions for 24 hours. Samples of the
liquid were then removed for mercury analysis by FAA.
PROCEDURE 12—SEDIMENT EXPERIMENTS
Two 1.0 g samples (wet basis) of sludge were reacted with 50 ml of
bromine water for 1 hour to oxidize all available mercury. The sludge
was then filtered and washed. Final filtrate and wash volume was 100 ml.
Aliquots were removed and mercury determined by FAA. Four 1.0 g samples
(wet basis) of sludge were added to four beakers, each containing 200 ml
of water at pH 6.2 (0.01 M phosphate buffer). Two of the mixtures were
stirred, the other two were unstirred. Samples of water were removed
after 48 hours and analyzed for mercury in the usual way.
In another experiment, four samples of fiber (1.0 ± 0.01 g) were placed
in the mixture of sludge (1.0 g) and water (200 ml), buffered to pH 6.2
with 0.01 M phosphate, and allowed to remain for eight days. The fibers
were rinsed with water to remove particulate and then reacted with 50 ml
of fresh bromine water for 5 hours. The solution was decanted into a
100-ml volumetric flask, the fiber was washed four times, and the rinses
were combined with the decanted liquid. Mercury was determined by FAA.
As controls, four fresh samples of fiber (1.00 + 0.01 g) were reacted
with bromine water and their mercury content was determined. The solu-
tion in contact with fiber and sludge was also analyzed for mercury.
49
-------
1
Accession Number
5
2
Subject Field & Group
05G
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
Arthur D. Little, Inc., Cambridge, Massachusetts
Title
WASTE WOOL AS A SCAVENGER FOR MERCURY POLLUTION IN WATERS
1 Q Authors)
Joseph P.
Tratnyek
16
21
Project Designation
16080HUBOU/72
EPA, WQO
Contract No.
68-01-0090
Note
22
Citation
23
Descriptors (Starred First)
*Water pollution treatment, *Heavy metals, *Textiles,
Bottom sediments, Adsorption, Sorption
25
Identifiers (Starred First)
*Water mercury removal, *Waste fiber utilization, *Wool fiber
27
Abstract
Laboratory studies demonstrated the feasibility of using waste wool and wool/polyester
blend fibers to remove mercury pollution from waters and bottom deposits. Nylon fiber
was shown to have limited potential. Within 24 hours, 90-95% of mercury at the 1-ppm
level was removed by the wool fiber. At higher levels of mercury, larger quantities
were removed, but the percentage decreased. Changes in pH (2 to 10) and temperature
(5 to 35°C) did not markedly alter efficacy of wool, nor did anaerobic conditions
or variation in water hardness. However, the presence of sulfide in water or sediment
reduced effectiveness of wool. Sources of mercury used were mercuric chloride, methyl
mercuric chloride, phenyl mercuric acetate, bis (2-methoxyethyl) mercury, and dissolved
metallic mercury.
Abstractor
Joseph Tratnyek
institution Arthur D. Little, Inc.
WR:102 (REV JULY 1969)
WRSIC
* U. S. GOVERNMENT PRINTING OFFICE : 1972—484-487/309
SEND TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
* GPO: 1969-359-339
------- |