INTERFERENCE MECHANISMS IN WASTE
STABILIZATION/SOLIDIFICATION PROCESSES
by
Larry W. Jones
Environmental Laboratory
U.S. Army Engineer Waterways Experiment Station
PO Box 631, Vicksburg. MS 39180-0631
Interagency Agreement No. DW219306080-01-0
Project Officer
' ~ Carlton Wiles
Waste Minimization, Destruction and Disposal Research Division
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
MSK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
I'.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
i UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
j WASHINGTON. D.C. 20460
OFFICE OF
RESEARCH AND DEVELOPMENT
SUBJECT: Transmlttal of OEETD Deliverable #4546, Interference
Mechanisms 1n Waste Stabilization/Sol 1d1f1cat1on
Process
FROM: Alfred 'f. ^
Acting Direct or, "^fVlce of "Envlron/nental
Engineering and Technology Demonstration (RD-681)
TO: Sylvia l.owrance
D1 recto?', Office of Solid Waste (WH-562)
The attached report entitled "Interference Mechanisms In Waste
Stabilization/Solidification Processes," was prepared by the Risk
Reduction Engineering Laboratory (RREL) of the Office of Environmental
Engineering and Technology Demonstration (OEETD). It 1s being
transmitted In fulfillment of our commitment to you 1n the FY 1988
Hazardous Waste Research Program planning process. This deliverable
supports the Agency's effort to Identify possible types of Interference
materials which might be encountered 1n typical waste/binder systems.
With this Information It may be possible to avoid undesirable effects.
This study 1nvest1gated the characteristics of Portland cement and
pozzolan, the effecrts of added constituents on their setting
characteristics, ard the effects of typical organic waste components on
the physical properties of the treated waste product. A glossary of
common cement terminology and three bibliographic appendices covering a
compilation of references and annotated citations for both Portland
cement and asphaltlc waste treatment systems Is also Included.
This report has also been forwarded to Alessl Otte of your staff for
review and utilization.
Attachment
-------
1
ORO CLEARANCE FORM
1. EPA Report No.
2. Series
3- Lab/Office Drift No.
HWERL-Sf-2086
* Copyright Permission
° Ye>
-------
NOTICE
The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency under Interagency Agreement
No. DW219306080-01-0 to the U.S. Army Engineer Waterways Experiment Station.
It has been subjected to the Agency's peer and administrative review, and it
has been approved for publication as an EPA document. Mention of trade names
or commercial products does not constitute endorsement or recommendation for
use.
-------
FOREWORD
Today's rapidly developing and changing technologies and Industrial
products and practices frequently carry with then the Increased generation
of materials that, 1f Improperly dealt with, can threaten both public
health and the environment. The U.S. Environmental Protection Agency Is
charged by Congress with protecting the Nation's land, air, and water
resources. Ihder a mandate of National environmental laws, the Agency
strives to fo-mulate and Implement actions leading to a compatible balance
between human activities and the ability of natural systems to support and
nurture life. These laws direct the EPA to perform research to define our
environmental problems, measure the Impacts, and search for solutions.
The Risk Reduction Engineering Laboratory 1s responsible for planning,
Implementing and managing research, development, and demonstration programs
to provide an authoritative, defensible engineering basis 1n support of the
policies, programs, and regulations of the EPA with respect to drinking
water, wastewater, pesticides, toxic substances, solid and hazardous wastes,
and Superfund-i-elated activities. This publication Is one of the products
of that research and provides a vital communication link between the research
and the user community.
This report describes and details the procedures used In the ultimate '
development of an "expert system" to evaluate closure options for RCRA
permitted waste management facilities. The report Identifies a consensus
need, the procedures used to select the language and delivery platforms
and the actual defining and development of the systems modules.
E. Timothy Oppelt, Acting Director
Risk Reduction Engineering Laboratory*
*Formerly the Hazardous Waste Engineering Research Laboratory (HWERL)
111
-------
ABSTRACT
The statilization/solidification of hazardous wastes Involves a series of
chemical treatment procedures. The waste is normally treated so as to complex
or bind the contaminants into a stable, insoluble for* (stabilization), or to
entrap the waste material in a solid and/or crystalline matrix (solidifica-
tion). Hazardous wastes contain many constituents that, could interfere with
the binding process. This project is concerned with identifying possible
interference Mechanisms between particular waste components and commercially
available waste-binding systems.
This report presents a literature review and information concerning
Portland cement and pozzolan chemistry, the effects of added constituents
(admixtures) on their setting characteristics, and the effects of typical
organic waste romponents on the physical and containment properties of the
treated waste |>roduct. These topics are presented so that conclusions may be
drawn as to possible types of interference materials that might be encountered
in typical wasV.e/binder systems. Also included are a glossary of common
cement terminology and three bibliographic appendices covering a compilation
of references and annotated citations for both Portland cement and asphaltic
waste treatment systems.
-------
Page
C. Bibliography: Effects of additives on
Portland cement and asphalt C-l
D. Glossary o:: cement chemistry terminology D-l
vi
-------
FIGURES
Number Page
1 Heat liberation curve by hydrating cricalcium
as determined by isothermal calorimetry 10
2 Changes in capillary pore space with degree of hydration 13
3 Possible adsorption mechanisms of organic
admixtures ?nto cement surfaces . 34
4 Influence of some organic admixtures on soluble CaO
and SiO. during early hydration of tricalcium
silicati 35
TABLES
Number
Page
1 Principal compounds of Portland cement
and their characteristics 9
2 Principal factors affecting strength of concrete
and stabilised/solidified waste 14
3 Influence of vater/cement ratio and degree of
hydration on concrete strength 15
4 Physical causes of deterioration of concrete
products ^
5 Crystallization pressures for salts l8
6 Durability to freeze-chaw cycles of neat cement
samples with different numbers of air bubbles 19
7 Stability of Portland cement solidified wastes to
high doses of gamma radiation 2l
8 Commonly used concrete admixtures 23
9 Some of the properties influenced by the use of
calcium chloride admixture in concrete . . . 28
10 Effect of accelerators trlethanolamina, calcium
formate, and calcium chloride on the rate of
hardening ami compressive strength of concrete 29
11 Compressive strength of mortars containing
retarders 3
30
12 Flexural strength of mortars containing retarders
13 Composition of simulated organic liquid waste used
by Clark et ill. (1982) . . . 39
vii
-------
ACKNOWLEDGMENTS
The study reported herein was conducted by the Environmental
Laboratory (EL) of the U.S. Army Engineer Waterways Experiment Station (WES)
under the sponsorship of the U.S. Environmental Protection Agency (EPA),
Hazardous Waste Engineering Research Laboratory, Cincinnati, OH. Author of
the report was Dr. Larry W. Jones. The report was edited by Ms. Jessica S.
Ruff of the WES '/.nfornation Technology Laboratory.
The study w«,a conducted under the general supervision of Dr. John
Harrison, Chief, EL; Dr. Raymond L. Montgomery, Chief, Environmental Engineer-
ing Division; and Mr. Norman R. Francingues, Jr., Chief, Water Supply and
Waste Treatment Group. Commander and Director of WES was COL Dvayne G. Lee,
CE; Technical Director was Dr. Robert W. Whalin.
The guidance and support of Mr. Carlton Wiles and Mr. Paul de Perein of
the Hazardous Waste Engineering Research Laboratory, EPA, are gratefully
acknowledged. Appreciation is also expressed to Mr. M. John Cullinane,
Mr. Mark Bricka, and Mr. Douglas Thompson of WES for reviewing the manuscript.
ix
-------
SECTION I
INTRODUCTION
Waste treatment processes that are designed to decrease the toxicity or
volume of hazardous wastes (e.g., precipitation or incineration) generate per-
sistent residues that must be prepared for safe final or ultimate disposal.
In most cases, final disposal of these residuals is by means of secure,
shallow-land burial. The disposal problem is most critical in the case of
wastes that cannot be destroyed or detoxified, such as heavy metal sludges and
brines. Thus, even with improved waste treatment methods, there will continue
to be a need fojr^ t»ehnft|oyy related to the safe land disposal of hazardous
wastes. /f^abili2ation/8olidificatjftn (S/S) of hazardous wastes before final
dlsposalHlSsTeen proposed as a"wiy to prevent release of the hazardous waste
constituents to the environment.
According to Title 40 of the Code of Federal Regulations, the U.S. Envi-
ronmental Protection Agency is responsible for evaluating the suitability of
hazardous waste for land disposal and for the examination of hazardous waste
delistlng petitions. A thorough understanding of the potential behavior of
stabilized/solidified wastes is necessary to make judgments as to the long-
term effectiveness of their containment. The extent to which various contam-
inants are securely held in stabilized/solidified wastes must be determined
for all S/S processes so that individual processes and delisting petitions can
be evaluated. There are several available methods for the S/S of hazardous
wastes. These methods have been developed primarily for inorganic wastes.
However, organic w.istes are increasingly being included as their volume con-
tinues to increase and S/S additives and technology improve. Some of the
chemical component*! of the complex wastes may interfere with the proposed S/S
process and cause tindesired results (e.g., flash set, set retardation, spall-
ing, etc.). This report surveys the literature concerned with the effects of
various waste constituents on the most common S/S binder systems: Portland
cement, pozzolan, and asphalt. Atchla time, there is_a paucityof quantita-
tive data concerning the snasific'effacts of theseTinteWerlttg~cgBpounds alone
or tn rnnrsrr itpniL,^rtlcularS7snpt:oceVs««.
LITERATURE SEARCH PROCEDURES
The literature search was conducted using facilities available in the
U.S. Army Engineer Vatervays Experiment Station (WES) Library and the WES
Concrete Technology Division Library. The search Included American Cement
Institute Journals and Indexes, American Society for Testing and Materi-
als (ASTM) Journals and Indexes, "Cement and Concrete Research" Journals,
"Chemical Abstracts," available textbooks, personal libraries and files of the
Concrete Technology. Division staff members, and telephone communication with
vafious experts in the field of S/S of hazardous wastes.
The Technical Information Center at the WES conducted two computer
searches using the Compendex Data Base, National Technology Information
-------
1
Service Data Base, and Technical Research Information Service Data Base using
the following key words! ' ~""""
additlveii organic waste
admixturii Portland cement
asphalt set time
asphaltic cement solidification
chemical solidify
chemical solidification stabilization
interference stabilize
organic waste
Additional computer searches were conducted using the Data Base
Index (DBI), Chemical Abstracts Data Base, Engineering Information and Techni-
cal Meetings Data Base (EIMET), Environment Abstracts Data Base (ENVIROLINE),
Electrical Power Information Abstracts Data Base (EPIA), and Geological Refer-
ences Data Base (GEOREF) of the American Geological Institute. The following
key words were usud:
admixtures interference
cement Portland cement
chemical interference solidification
hazardous waste stabilization
Three bibliographies developed through these procedures are included as
appendices. Appendix A is an annotated bibliography of references on effects
of additives on Portland cement concrete, listed alphabetically by first key
word in the title. Appendix B is an annotated bibliography of references on
the effects of additives on asphalt materials, arranged similarly to Appen-
dix A. Appendix C is an alphabetical listing of all applicable references, by
author's name.
A glossary of cement chemistry terminology is presented as Appendix D.
EFFECTS OF SPECIFIC MATERIALS ON CEMENT AND POZZOLAN CHEMISTRY
Although cheolcal stabilization and solidification of hazardous wastes
before disposal is increasing in importance, very little work has been done
concerning the effects of specific waste components on the physical and con-
tainment properties of the final S/S waste product. Many categories of
inorganic and organic materials are known to have either beneficial or harmful
effects on concrete and asphalt products as used in the construction industry.
Such admixtures (oaterials added to change the physical properties of con-
crete) are routinely added to concrete formulations to retard set, lower heat
production, increase water resistance, and/or increase final product strength.
Likewise, care has to be taken to see that aggregates or other components do
not contain deleterious materials^
A review of cement literature reveals that the chemistry of the hardening
of concrete is not well understood although the basic chemical reactions and
their products are fairly well known. Several theories have been proposed and
are discussed concerning how additives produce their effects by interfering
with cement hydration reactions, but none appears to be entirely satisfactory.
-------
Pozzolan cheiiistry is less well known, although it is thought to be analogous
in some ways to that of Portland cement.
The complexity of the hydration reactions and the effects of slight vari-
ations in process formulations or mixing parameters complicate the prediction
of the effects of additives on the final S/S waste product. Effects of
single-component additions are only poorly understood. Predicting the effects
of complex and variable compositions such as those found in a real waste is
beyond the current state-of-the-art.
As an example, the use of short-term or accelerated curing tests for cop-
pressive strength may give erroneous results as to the long-term stability of
the binder-waste mix. Materials that cause the retardation of set often bring
about increased -long-term strength in both Portland cement and lime/fly ash.
,S/Ssy_stej|i»-JFor mixes containing organic wastes or other known set
~refarders7'a aibriav cure,(inAddition to 28-day samplesMs recommended. This
long cur1ng/t:fme brings about~bbv1dus difficulties InthY day-to-day operation
of a waste S/S plant. No short-term testing protocols have yet been developed
to adequately predict long-term stability, durability, and strength
characteristics of S/S waste products. :
Nonpolar organic materials of low volatility should not hinder strength
Sdevelopment in cement or pozzolan systems. Since chlorinated hydrocarbons
comprise a large proportion of the hazardous organic wastes, a study of their
effects upon these systems is strongly recommended. Organ1c_wastes that
A contain hydroxyMejAM sugars and alcohols)_ or carboxyHc~acid (e.JU» citric
* .-*-,J- --^^ -- /--- i- aBines) can-6t-expected Jo delay, or
stop altogether the pozzolanic reactions. These functional groups are present
in wastes such as biological sludges, paint sludges, and Jiany
The final properties of a waste-cement mixture are highly dependent upon:
1) the constituents of the waste that may interfere with the setting reac-
V ,/V°ns» 2* tne water/cement ratio which is the primary determinant of the size
,\reconfiguration of the void space and permeability of the final product;
* yK3) presence of admixtures such as fly ash, surfactants, or others that may
.modify setting or final strength parameters of the product; and 4) procedures
, y*nd conditions used for mixing and curing the treated wastes. The strength .
$ and durability of concrete products are directly related to the type and
'' number of voids in the final product. Adequate design of the formulation and
v , mixing parameters can change the durability/accessibility of water to
y '.£*,.waste constituents of the final S/S product by several orders of magnitude.
Asphaltic and polymeric materials used for waste S/S also are affected by
low levels of ;;om« waste components. Solvents and oils may weaken and liquefy
the asphaltlc base. Oxidizers and some inorganic materials may cause or
promote(f^feS) or flash-set of the final mixture. Care should be taken to
thoroug!Hy--tes1: specific wastes that are candidates for this treatment
process. '
STUDIES OF HASTE INTERFERENCE WITH S/S PROCESSES
A great varfe^ty 1n the types of reagents used and testing performed is
found 1n the fiirpublished studies which allude to the effects of waste
-------
components on the S/S process. Details of materials and reagents, preparation
and mixing procedures, and product testing protocols are often sketchy and
incomplete. In this review, all reports which Impinge upon effects of dif-
ferent waste constituents on specific S/S processes are discussed in the light
of current testing and disposal requirements* No broad generalizations could
be gleaned from the studies reviewed, but a list of thosevmateriala which -'''-
should be studied further include heavy metal/interferences such as lead, ' ^
copper, and zinc; organics, such as grease and oils, phenolics, degreasers,
and pesticides; and stringent conditions such as high sulfate and strong
bases. ';" ;v '" " ' "' - '
There is a clear need for further experimental work to obtain physical
and chemical data relative to cementitious and asphaltic-S/S treatment ays- -
terns. Enough bapic information should be developed so that-valid prediction
and modeling of waste-binder interactions are possible. -This ability would >'
overcome the current need to test each specific waste-binder combination for ''-
possible interferences. However, the variability and complexity of most-waste
streams, as well as cement and pozzolanic setting reactions, still
elude broad generalizations. ***'' '*&.';-&,?
;-T < .
~*4' " J "
. I.. .
**- . ' ." ..
>'.~" _ - _-^.^.C- 'w
. -j -_n r..< .-.,;
,r '. .* l»»'
-------
SECTION 2
CONCLUSIONS AND RECOMMENDATIONS
Although chemical stabilization.and solidification (S/S) of hazardous
wastes before disposal is increasing in importance, very little work has been
done concerning the effects of specific waste components on the physical and
containment properties of the final S/S waste product. Many categories of
inorganic and organic materials are known to have either beneficial or harmful
effects on concrete and asphalt products as used in the construction Industry.
Such admixtures (materials added to change the physical properties of con-
crete) are routinely added to concrete formulations to retard set, lower heat
production, increase water resistance, and/or increase final product strength.
Likewise, care has to be taken to see that aggregates or other components do
not contain deleterious materials.
A review of cement literature reveals that the chemistry of the hardening
of concrete is not well understood, although the basic chemical reactions and
their products are fairly well known. Several theories have been proposed
concerning bow additives produce their effects by interfering with cement
hydratlon reactions, but none appears to be entirely satisfactory. Pozzolan
chemistry is less well known, although it is thought to be analogous in some
ways to that of Portland cement.
The complexity of the hydration reactions and the effects of slight vari-
ations in process formulations or mixing parameters complicate the prediction
of the effects of additives on the final S/S product. Effects of single-
component additives are only poorly understood. Predicting the effects of
complex and variable .££££»£§, compositions such as those found in a real
waste is beyond the current state of the art.
As an example, the use of short-term or accelerated curing tests for com-
pressive strength may give erroneous results as to the long-term stability of
the binder-waste mix. Materials that cause the retardation of set often bring
about increased long-term strength in both Portland cemanc and lime/fly ash
S/S systems. For mixes containing organic wastes or other known sec
retarders, a 90-day curs (in addition to 28-day samples) is recommended. This
long curing time brings about obvious difficulties in ths day-to-day operation
of a waste S/S plane. No short-term testing protocols have yet been developed
to adequately predict long-term stability, durability, and strength charac-
teristics of S/S waste products.
Nonpolar organic materials of low volatility should not hinder strength
development in cement or pozzolan systems". Since chlorinated hydrocarbons
comprise a large proportion of the hazardous organic wastes, a study of their
effects upon these systems is strongly recommended. Organic wastes that con-
tain hydroxyl or carboxylic acid groups (and perhaps amines) can be expected
to delay or stop altogether the pozzolanlc reactions. These functional groups
are present in wastes such as biological sludges, paint sludges, and many
solvents.
-------
The final properties of waste-cement mixtures are highly dependent upon:
1) the constituents of the waste that may interfere with the setting re*c- j
tions; 2) the water/cement ratio, which is the primary determinant of the size
and configuration of the void space and permeability of the final product;
3) presence of admixtures such as fly ash. surfactants, or others that may
modify setting or final strength parameters of the product; and 4) procedures
and conditions used for mixing and curing the treated wastes. The strength
and durability of concrete products is directly related to the type and number
of voids in the final product. Adequate design of the formulation and mixing
parameters can change the durability and accessibility of water to waste con-
stituents of the final S/S product by several orders of magnitude.
Asphaltic and polymeric materials used for waste S/S also are affected by
low levels of some waste components. Solvents and oils may weaken and liquefy
the asphaltic base. Oxidizers and some inorganic materials may cause or pro-
mote fires or flash set of the final mixture. Care should be taken to
thoroughly test specific wastes that are candidates for this treatment
process.
A great variety in the types of reagents used and the testing performed
is found in the few published studies which allude to the effects of waste
components on the S/S process. Details of materials and reagents, preparation
and mixing procedures, and product testing protocols are often sketchy and
incomplete. In this review, all reports that Impinge upon the effects of dif-
ferent waste constituents on specific S/S processes are discussed in light of
current testing and disposal requirements. Mo broad generalization's could be
gleaned from the studies reviewed, but a list of those materials which appear
to be appropriate for further study include heavy metals interferences, such
as lead, copper and zinc; organics, such as grease and oils, phenolics,
degreasers, and pesticides; and stringent conditions such as high sulfate and
strong bases.
There is a clear need for further experimental work to obtain physical
and chemical data relative to cementitlous and asphaltlc S/S treatment sys-
tems. Enough basic information should be developed using standard mixing and
testing protocols so that valid prediction and modeling of waste-binder
interactions are possible. The ability to generalize concerning interference
effects would overcome the current need to test each specific waste-binder
combination for possible interferences. However, the variability and
complexity of moat vast* streams as well as cement and pozzolanic setting
reactions may still preclude such generalizations.
-------
SECTION 3
CEMENT AND POZZOLAN PROPERTIES THAT
MAY AfFECT WASTE CONTAINMENT
INTRODUCTION
The ability of a atabilized/solidifitd (S/S) waste product co contain a
givtn hazardous constituent depends primarily upon its resistance to leaching
or volatilization of waste constituents and its long-term durability. This
section develops the background necessary for an understanding of factors con-
trolling leaching and durability characteristics of cement and pozzolanic
systems.
Comparison of the finished S/S waste product with standard construction
materials-is out tntlge4y-Anpropria\te_ since^moat-hararrdous wastrsTreated by
S/S technologies are liquid slurries with relatively low solids content (10 to
40 percent .solids by weight). The mixture of a solidification agent such as
cement with a slurrieoT'VastB more closely resembles a hydra ted cement -paste
rather than a typical concrete with a large proportion of aggregate (60 to
80 percent of the final concrete volume). Experimentation with hydrated
cement pastes provides a basis for speculation on the effects of various haz-
ardous waste constituents on a S/S waste product.
As noted above, the characteristics of S/S wastes that are most Important .ci
to waste containment are Reachability gntl f<"T'*MrLJry However, unconfined
compressive strength (UCS) is the"commonly specified measure of concrete
quality since it is determined easily and inexpensively by a reproducible
test. No correlation has been demonstrated between UCS and leachability.
Cote et al. (1984) has based a theoretical model of the leaching of spe-
cific, low-to-high solubility material from a solid matrix to a large extent
on the rate of internal diffusion of the material in the solid matrix.
Internal diffusion of soluble ions or molecules in a solid matrix depends
largely upon the relative amount and character of void spaces in the solid,
which in turn relates directly to the permeability of the porous solid and its
ultimate strength.
Numerous researchers have Investigated the chemistry and the physical
characteristics of cement and pozzolanic mixtures. These studies serve, as an
excellent foundation for understanding the similar properties of cement- and
pozzolan-based S/S processes. This section presents a brief overview of
cement and pozzolan chemistry and terminology. Detailed information on the
subject is available in the following primary sources.
1. Bye, G. C. 1983. Portland Cement; Composition, Production and
Properties. Pergamon Press, New York. 149 pp.
-------
2. Mehta, P. K. 1987. Concrete Structura, Properties and Materials.
Prentice-Hall. lac., Englewood Cliffs, NJ. - ' -
3. Neville, A.M. 1981. Properties of Concrete. Pitman Publishing, Inc. , .
Marshfield, MA.
4. Portland Cement Association. 1968. Design and Control of Concrete
Mixtures, llth ed., .Portland Cement Association, Skokie, IL.
5. Powers, T. C. 1958. The Physical Structure and Engineering Properties of
Concrete. Bulletin 90, Portland Cement Association, Skokie, IL. - ~
6. U.S. Bureau of Reclamation. 1975. Concrete Manual. Washington, DC.
7. Wood*, H. 1968. Durability of Concrete. ACI Monograph 4. American
Concrete Institute, Detroit, MI.
PORTLAND CEMENT
Portland cement is probably the most common cementitious additive for the
S/S of hazardous waste. Portland cement is highly uniform, relatively
inexpensive, and widely available; it can be worked in normal construction
machinery by unskilled labor; and it produces products with highly predictable
characteristic*. A variety of standard types of Portland cemenc arc available
for specific applications, and many admixtures have been developed to modify
its setting and final properties.
The Solids in Hydrated Cement
The hydration of Portland cement is a series of simultaneous and consec-
utive reactions between water and solid cement constituents which occur in the
setting and hardening processes. Anhydrous Portland cement consists of angu-
lar particles (usually 1 to 50 urn) with a chemical composition of the primary
clinker materials that corresponds approximately to C,S, C.S, C.A, and C.AF,
] where C - CaO, S - SiO., A - Al 0-, F - Fe 0-. S - SO* an* H -%0. In ordi- "7
*-nary Portland cements, the respective amount! range between 45 and 60. 15 and -*
30, 6 and 12, and 6 and 8 percent (with a small added amount of CS). Table 1
lists selected characteristics of these materials.
Dispersing Portland cement in water causes the various constituents to go
into solution, which rapidly saturates the liquid phase with the various ionic
species. The first needle-like crystals of calcium sulfoaluminate hydrate
(ettringite) appear within a few minutes of cement hydration (see Equation 2
below). After a few hours, largt, prismatic crystals of calcium hydroxide
(CH) and very small fibrous crystals of calcium silicate hydrates (C-S-H)
begin to fill the empty spaces formerly occupied by water and the dissolving
cement particles. The most important reactions in later acages are the hydra-
tion of calcium silicate's (to produce C-S-H), which continues for manyTgbntha
(Popovics " "
The rapid changes in the solution composition which occur upon initial
hydration are followed by a relatively constant composition, depending upon
the initial cement material and the water/cement ratio. The solution is
-------
TABLE 1. PRINCIPAL COMPOUNDS OF PORTLAND CEMENT AND THEIR CHARACTERISTICS
Approximate
composition
Abbreviated
formula
Common name
Principal
3CaO«S102 B2CaO-Si02
c3s sc2s
Alit« Belite
^MgO, AUCu, $gO> Al,ol (i
23, a 2 3 e2 3
C3A C4AF
Ferrite,._phase, Fss
10,,(Mid>\ (^-SiO,,^,^
impurities
Common crystal-
line form
Proportion
present (£)
Range . ... 35-65
Average in
ordinary
ceaent 50
Rate of reaction
with water Medium
Contribution to
strength
Early age Good
Ultimate Good
Heat of hydration Medium
Typical (cal/g) 120
nic Monoclinic Cubic, Orthoftreaisic
orthorhombic
10-40
25
Slow
Lov
High
Lov
60
0-15
Fast
Good
Medium
High
320
5-15
8
Medium
Good
Medium
Medium
100
supersaturated in calcium hydroxide (CH) and hydrated calcium sulfate (CSH)
until the reaction involving the formation of calcium sulfoaluminate removes
most of the calcium sulfate from solution. The CH, however, remains in solu-
tion, and solid CH is present at all stages of the cement setting reactions.
Since the hydration of trlealclum silicate (C.S) is the dominating mech-
anism in the overall hydration process, several distinct stages in its hydra-
tion have been identified (Rondo and Daloon 1969; Young 1972, 1976). These
stages are shown in the conduction calorimetry curve in Figure 1. During
Stage I, the initial hydrolysis of C.S occurs, with a rapid release of heat
and calcium hydroxide into solution. This leaves a C^S particle with an
approximately 10-nm insulating rind of hydrated C.S.
In Stage II the release of CH continues, but at a slower rate limited by
the ability of CH to interact with available water through the rind. The
-------
TOM
Figure 1.
Heat liberation" curve by hydratlng tricalcium silicate as
determined by isothermal calorimatry (Sourca: Kondo and
Daimon 1969).
thin layars of hydracad calcium ailicaca (C-S-H) that maka up tha rind ara
probably a polymerization of tha hydrolyzad allicata groupa. Staga III ia
raprasantad by an accalaratad C.S cryetallization and tha initiation of cal-
cium hydroxida crystallization. Simultaneously, a -tranaformation of tha C-S-H
gel occura, expoaing additional surface area of tha anhydroua particle.
Deceleration of thaaa diffusion-controlled reactiona typifiea Stages IV and V,
which continue aa slowing exothermic reactions.
Hydration of tricalcium aluminate (C.A), while not being critical to
strength development, is crucial in determining cement see time and other
Stage I and II characteristics (Hansen 1959). Hydration of C.A in the absence
of suifiim ions can be represented by the following equation:
21 H-
slow
(matastable hexagonal hydrates)
!C3AHfi + 9H
(stable hydrate)
(1)
The heat liberated in Equation 1 la sufficient to apeed up tha reaction to the
stable form, C.AH,. Portland cement does, however, contain a small amount of
aulfata as calcium sulfata (gypsum). This alters the hydration to the form
of
3CSH
26H-
(2)
(ettringite)
The addition of 26 water moleculea during tha formation of ettringita brings
about a correaponding volume increaar. When ettringita forms early in the
solidification proceas, tba additional space ia not a problem; it even devel-
ops binding properties that contribute to the early strength of tha product.
This phenomenon can actually be used to bind large amounta of sulfate in a
solidification product by using clay-rich (alumina) binders. However, if the
ettringite forms after solidification, the additional space required loosens
the structure and causes an observable increase in permeability with a
10
-------
concomitant lose in strength (Wiedemann 1982). Afttr a period of time, Che
length of which depends on the alumina-to-sulfate ratio of the cement, the
ettringite crystal* may react with C.A according to
4H + 2C2A + C3A'3CS-H32 - *-3C3A'CS'H32 + 4H (3)
(ettringite) (monosulfoaluminate)
Monosulfoalualnate in Portland cement concrete makes the concrete vulnerable
to sulfate attack.
The major constituents of fully hydrated cement paste are discussed
below.
1. Calcium silicate hydrate (C-S-H) makes up 50 to 60 percent of the
volume of solids in fully hydrated Portland cement paste. Termed a
C-S-H gel in older literature, this ill-defined compound forms small,
poorly defined crystalline fibers in a retlcular network.
2. Calcium hydroxide crystals constitute 20 to 25 percent of the volume
of solids in the fully hydrated paste; it tends to form large crys-
tals with distinctive hexagonal-prism morphology. The presence of
significant amounts of calcium hydroxide makes the concrete reactive
to acidic solutions.
3. Calcium sulfoalumlnates make up 15 to 20 percent of the solids volume
in the fully hydrated paste but appear to play a minor role in the
structure and properties of the final product.
Voids in Hydrated Cement and Concrete
The several kinds of voids in hydrated cement paste have great influence
on its final properties of strength, durability, and permeability. The
smallest voids, which occur within the C-S-H gel structure, are 0.5 to 2.5 am
in diameter. They account for about 28 percent of the porosity in solid
C-S-H. These small voids have little effect on the strength and permeability
of the final product but appear to be important in drying shrinkage and creep.
Capillary voids account for the larger spaces which are not filled by
solid components. In well-hydrated , low water/cement ratio mixes, capillary
voids range from 10 to 50 nm, but in high-ratio mixes they may be as large as
3,000 to 5,000 nm. It is generally held that pore size distribution, and not
simply total capillary porosity, is a' better criterion for evaluating the
characteristic of a cementitious product. Capillary voids larger than about
50 nm are thought to be detrimental to strength and permeability, while voids
smaller than 50 nm are more important to drying shrinkage and creep. Capil-
lary voids limit the strength of concrete by acting as "stress concentrators"
(Hirsh et al. 1983).
The third type of voids, usually called "air voids," are generally spher-
ical and usually range from 0.05 to 0.2 mm but may range up to 3 mm. Air
voids are usually introduced intentionally into the hydrated cement paste to
increase the resistance of the final product co freeze-thaw (frost) damage
even though they typically adversely affect its strength and permeability.
11 _
-------
Depending on the environmental conditions, Che voids are capable of hold-
ing large amounts of water. Capillary water (in voids 5 ma or larger) is bulk
water that is largely free from attractive surface forces. Water in voids
greater than about 50 nm is considered free-water since its loss causes no
shrinkage in the final product, while loss of water held by capillary tension
in voids from about 5 to 50 am may cause some shrinkage. Changes in the
amount of capillary pore space with degree of hydration are illustrated in
Figure 2a for a c«m*nt paste with a water/cement ratio of 0.66 and in Fig-
ure 2b for fully hydrated pastes of different water/cement ratios.
Absorbed water is close to the surface (probably within l.S no of the
surface) and held by hydrogen bonding and Van der Wai forces. Loss of
absorbed water, even in air of 30-percent relative humidity, is mainly respon-
sible for the shrinkage and cracking of the solidifying mas*. The water more
tightly bound in the interlayers of the C-S-R structure will be lost only in
air with relative humidities below about 10 percent. The loss of water from
the C-S-H structure will cause considerable drying shrinkage.
POZZOLANIC MATERIALS
By definition, pozzolans are materials that display no cementing actions
by themselves but which contain constituent* that combine with lime at ordi-
nary temperatures and in the presence of water to fora cementltious compounds.
At present, the main pozzolans used commercially are fly ash from coal-fired
power plants and kiln dust from lime or cement kilns. Fly ash is the fine ash
produced by combustion of powdered coal with forced draft. The ash is carried
up in the flue gases and collected in special equipment such as electrostatic
precipitators or bag houses. Fly ash was first reported in the literature in
terms of its pozzolanlc properties in 1937 (Davis et al. 1937).
Most traditional literature is concerned with fly ash as an admixture in
cement rather than as an independent potzolanic material. Minnick (1967)
investigated eight U.S. fly ashes and their reactions with hydrated lime over
reaction periods up to 32 weeks. Minnick has suggested pozzolanic reactions
according to the following:
H0
H
(C-S-H of varying stoichiometry)
H,0
CH + A . * > cxAyHz (5)
(hexagonal and cubic eliminate hydrates)
HO
CH + A + S **-CxAySzHv (6)
(hydrogarnets)
H,0
CH + S + A ^**cxAy(C§)zHw (7)
(ettringite and derivatives)
12
-------
300 r
LEGEND
| 1 CAPILLAAV KJMS
ANHYDROUS CEMCNT
HYOftATION FMOOUCT
DAYS
HYDRATED
HYDRATION
DEGREE
NONE
7d
50%
28d
75%
1 yr
100%
Figure 2a. Change in the amount of capillary pore space with degree of
hydration for a cement paste with a 0.66 water/cement ratio
(after Mehta 1987)
LEOtNO
I
300
ISO
JOO
P8
HATKJ
Figure 2b.
Change in the amount of capillary pore-apace for a fully
hydrated cement paste with differing water/cement ratios,
based on equal .amounts of Portland cement (after
Mehta 1987)
13
-------
The abov« reactions yield products whose properties are similar to the reac-
tion products of Portland cement. One primary difference is that pozzolanic
reactions consume lime rather than produce it, as with typical Portland cement
hydration. This is an important factor in sulfate corrosion resistance and in
differences in the effects of additives on the two S/S materials.
Pozzolanic reactions, while not identical, are similar to Portland cement
reactions. The strength of products produced in both processes results from
the formation of hydrated calcium silicates (C-S-H); however, pozzolanic reac-
tions are not pure hydration reactions like those of Portland cement. Port-
land cement involves two or more solid phases plus water. ,Pozzolanic
reactions generally art much slower than cement reactions, and set times are
usually measured in days or weeks instead of hours.
THE STRENGTH-POROSITY RELATIONSHIP
Factors that influence the strength of a cement on pozzolanic mass are
listed in Table 2. By far the most important factor is the water/cement
ratio-porosity relationship. This is usually explained as the natural conse-
quence of the weakening of the cement matrix caused by increasing porosity
with increasing water/cement ratios (as discussed above, see especially Fig-
ure 2b). Typical data illustrating this point are shown In Table 3. Fig-
ure 2b also illustrates the increase in strength with increasing hydration
(time of cure) which corresponds to a decrease in the total pore space.
TABLE 2. PRINCIPAL FACTORS AFFECTING STRENGTH OF CONCRETE AND
STABILIZED/SOLIDIFIED WASTE
Matrix Porosity
Water/cement ratio
Degree of hydration
Curing time
Temperature
Humidity
Mineral admixtures
Air content
Specimen Parameters
Geometry
Dimensions
Moisture
Chemical Interactions with Wast*
Wast* constituents
Cement aad vaste loading
Stress Parameters
Stress type
Rate of application
14
-------
TABLE 3. INFLUENCE OF WATER/CEMENT RATIO AND DEGREE
OF HYDRATION ON CONCRETE STRENGTH
Water/cementUnconfined compresslve strength (psi)
ratios after different lengths of cure __
by weight 1 day 3 days .7 days 28 days
0.35
0.45
0.55
0.65
0,75
1,300
900
550
300
200
2,900
2,200
1,650
1,200
950
4,450
3,500
2,750
2,000
1,550
6,300
5,250
4,350
3,450
2,800
Source: Portland Cement Association' 1968.
Notes: Cylinders (6 by 12 in.) (15 by 30 cm) of normal type I cement
undergoing moist-cure. To convert pounds (force) per square inch to
megapascals, multiply by 0.006894757.
Water/Cement Ratio and Degree of Hydration
The relationship between porosity and strength is of great importance to
the prediction of the ability of a given waste-cement mixture to contain the
contaminants when subjected to various environmental conditions. Therefore,
compressive strength may be an indicator of the S/S waste's resistance to
leaching. This thought is further developed below in the section on
durability-permeability.
Mineral Admixtures
The use of pozzolanic and cementitious by-products such as fly ash, blast
furnace slag, or kiln dusts as admixtures is an important issue in waste S/S.
When used in addition to, or as a partial replacement for Portland cement, the
presence of the mineral generally retards the rate of strength gain. However,
the mineral admixture reacts with the excess calcium hydroxide present in the
hydrated Portland cement to form additional C-S-H and leads to a significant
reduction in porosity and ultimata strength gain in the final product (Marsh
et al. 1985). Consequently, considerable improvement in the ultimate strength
and impermeability (water tightness) can be achieved by their incorporation.
Air EntraiTtw*nt
Additives causing stable air incorporation into the cement paste are uni-
versally deleterious to the ultimate strength and impermeability of the con-
crete, due most likely to the added large-pore space. However, the entrained
air increases the resistance of the products to freezing.
Specimen Parameters
The standard test specimen' for the uniaxial compression test in the
United States (ASTM C 469) is a right cylinder of 6 in. (15 cm) diameter (d)
-------
and 12 in. (30 cm) height (h) (ratio h/d - 2 ). Measured strengths vary
indirectly with h/d ratio, with a h/d ratio of 1 giving 15 to 20 percent
higher strength and a ratio of 4 about 10 percent less. Specimens of dif-
ferent sizes, but having a h/d 2 , have different measured strengths. A
3- by 6-in. (7- by 15-cm) cylinder has about 6 percent higher strength, and a
9- by 18-in. (23- by 46-cm) specimen has about 10 percent lower strength than
the standard 6- by 12-in. cylinder.
Strength tests based upon the 6-in. standard test cube, which are common
in European countries, generally indicate strengths 10 to 15 percent higher
than tests of the same material using 6- by 12-in. cylinder. Similarly, the
use of a 2-in. cube for compressive strength testing would be expected to give
15 to 25 percent higher strength values for the same cement mixtures.
Concrete products also perform differently under different stresses, com-
pressive strength being the outstanding feature of structural concrete. Most
designs emphasize compressive strength, since the tensile strength of concrete
is usually an order of magnitude lower. The rate of stress application is
also important, as long-term stress brings on creep strain due to moisture
movement in the C-S-H voids and the development of microcracka.
Chemical Interaction of Waste with the Cement Binder
Wastes most amenable to S/S are usually water-based sludges containing a
wide variety of inorganic and organic constituents at different and varying
concentrations. Predicting the effect of these poorly defined mixtures on the
strength of hydrating cement paste without benefit of bench- or pilot-scale
studies is complex at best and probably beyond the state of the art.
THE DURABILITY-PERMEABILITY RELATIONSHIP
Long-term durability of the S/S waste product is a prime consideration in
the design and specification of waste S/S systems. Prediction of ^nj:"
integrity of the final waste fprm require, consideration of all possible modes
of failure (Table 4). For cementitiou. S/S products, water is generally
involved in every form of deterioration; in porous solids, permeability of the
material to water usually determines the rate of deterioration. Internal
movement and changes in the structure of water are known to cause "«?<£
volume change, of mmny type.. Example, ar. freezing of water into ice, forma-
tion of ordered structure of water inside fine pores, development of osmotic
pressures due to different ionic concentration., and hydrostatic pressure
buildup by differential vapor pressure.. All of these can lead to large
internal stre.... within a moist solid- and result in its ultimate failure.
In porou. solid., water al.o act. a. a vehicle for transport of
through the material, both aggressive ion. into and wa.t. ma «ri.l» out
discussion of permeability is therefor, of ba.ic interest to *»^«"
of both durability and l.achability. Permeability of concrete ?»*»»
depends primarily on the water /cement ratio (which ^'ermines th? size, vol
ume. and continuity of capillary voids) and the development of *«""
that occur between' the cement past, and the surface
ta occu s
as aggregates or waste solids) . The suspended particulars in waste wd|»«
Uctlng as small aggregate) are typically very small. In general, the smaller
16
-------
TABLE 4. PHYSICAL CAUSES OF DETERIORATION OF CONCRETE PRODUCTS
Surface Wear
Abrasion
Erosion
Cavitation
Cracking
Volumt changes
Moisture gradients and humidity
Crystallization pressures of salts in pores
Exposure to temperature extremes
Freeze-thaw action
Fir«
Structural loading
Overload or impact
Cyclic loading
the suspended particles, the fever the microcracks at their surface, and the
lower the overall permeability of the final product. The primary considera-
tion of permeability of S/S waste sludge* of small particle size then, is the
water/cement ratio.
Cracking by Crystallization of Salts in Pores
Stabilization/solidification waste products often contain substantial
amounts of salts and/or organic molecules with appreciable water solubilities.
Concentration of these materials at or belov the surface of the solid where
evaporation of pore water la occurring can cause the development of super-
saturated solutions and the formation of salt crystals in the pores of the S/S
product.
Crystallization occur* only when the concentration of the solute (C)
exceeds the saturation concentration (Cs) at a given temperature. Generally,
the higher the degree of supersaturation (the ratio of C/Ca ), the greater
the crystallization pressure exerted on the solid structure. Table 5 presents
crystallization pressures for C/Ca 2 of a series of salts that are common
in stone and concrete materials. These values were calculated by Uinkler
(1975) in an effort to understand the rapid deterioration of stone and con-
crete monuments by snog and acid (high-sulfate) rain. For example, at
C/Cs - 2 , halite (NaCl) at 25* C produces 605 atm (61 MPa) of pressure,, and
at C/Cs 10 , 2,020 atm (205 MPa). These pressures are strong enough to
disrupt the structure of the stone or concrete products which contain these
constituents. As seen in Table 5, common salts have a wide range of
crystallization pressures. Due to the wide variety of materials found in
hazardous wastes, this effect can be appreciable.
Damage typical of this effect is the powdering or spalling of the subsur-
face of the solid material which progressively deepens into the material as
17
-------
TABLE 5. CRYSTALLIZATION PRESSURES FOR SALTS
Pressure (atm)*
C/C - 2
Salt
Anhydrite
Bischofite
Dodekahydrate
Epsomite
Gypsxun
Halite
Heptahydrite
Hexahydrite
Kieserite
Mirabilite
Natron
Tachhydrite
Thenardite
Thermonatrlte
Chemical Formula
CaS04
MgCl2
MgS04
MgS04
CaS04
NaCl
Na2C03
MgS04
MgS04
N.2S04
N.2C03
2MgCl2
Na2S04
Na2C03
6H20
12H20
7H20
2H2°
7H20
6H20
H20
IOH20
10H20
CaClj 12H20
H20
0" C
335
119
67
105
282
554
100
118
272
72
78
50
292
280
50s C
398
U2
80
125
334
654
119
141
324
83
92
59
345
333
Source: Winkler (1975).
* To convert atmospheres (standard) to megapascals, multiply by 0.101325.
its porosity increases. Damage to S/S products due to wet-dry cycles may be
to a large extent due to the cyclic dissolution and crystallization of con-
tained salt.
Wet-Dry Cycling
Other than cracking by salt crystallization as described above, vet-dry
cycling of normal concrete products does not usually produce significant
damage to its structure. However, as the total proportion of cement is
reduced or the water/cement ratio is increased, as is common in waste S/S
practice, vet-dry cycling may cause rapid deterioration of the S/S waste
product. Jones and Malone (1982) reported rapid deterioration of S/S
inorganic waste products produced by commercial S/S vendors using ASTM
standard test procedure D559-57 for compacted soil-cement mixtures (ASTM
1976).
18
-------
Effects of Radiation
Large doses of gamma radiation do not appear to affect setting properties
or cause appreciable loss of strength or increased leachability in Portland
cement solidified waste products (Neilson et al. 1987). Waste forms produced
with 15.6 wtZ, fully loaded (with radioactive cesium and strontium) resin
beads, 21.7 percent water, and 62.7 percent Portland type I-II cement were
subjected to high doses of gamma irradiation (4.0 to 5.7 * 10 rads) and sub-
jected to testing. Results of the tests are presented in Table 7.
TABLE 7. STABILITY OF PORTLAND CEMENT SOLIDIFIED WASTES
TO HIGH DOSES OF GAMMA RADIATION
Attribute
Waste*
Control
Irradiated
(5 * 108 rads)
Compressive strength
Leachability index**
134 J 137
for ""Ca and
(averaged)
Distilled water
Distilled water
Sea water
Sea water
Cs
PF-7
PF-24
PF-7
PF-24
PF-7
PF-24
20.200 t 3,300
25,000 ± 5,000
25,100 ± 9,900
22,800 ± 11,800
10.3
10.5
9.6
10.4
9.35
10.0
10.0
10.9
* "Fully loaded resin beads from prefilter no. 7 or no. 24.
** Leachability index is negative log of leaching rat* (larger numbers leach
less) (see American Nuclear Society (ANS) 1986).
21
-------
1
SECTION 4
EFFECTS OF STANDARD ADMIXTURES ON PORTLAND CEMENT/POZZOLAN PRODUCTS
INTRODUCTION
The properties of concrete in both the fresh and hardened state are typi-
cally modified by adding specific materials to the concrete mixtures. The
additives, generally called admixtures, vary widely in chemical composition
and may modify more than one property of the concrete mixture. It has been
estimated that 88 percent of concrete placed in Canada and 70 percent placed
in the United States includes one or more admixtures (Mehta 1987).
An understanding of the effects of common, well-studied admixtures gives
a basis for assessing the possible range of effects.of the S/S waste constit-
uents on the concrete or pozzolanlc matrix. American Concrete Institute (ACI)
Committee 212 report Admixtures for Concrete (1963) contains a thorough report
on the various effects of standard admixtures in concrete production, and ASTM
Standard C 494-82 covers chemical admixtures for concrete (ASTM 1982). Vivian
(1960) has written an excellent review of early work in this field. Rixon
(1978), Schutz (1983). and Ramachandran (1984) have written more recently on
chemical admixtures.
Admixtures fall broadly Into three categories: 1) surface active mole-
cules that work on the cement-water system immediately upon addition by influ-
encing the surface tension of water and by absorbing onto the surface of
cement particles; 2) set-controlling materials that.ionize and affect the
chemical reaction between ths cement and the water only after several minutes
or hours; and 3) finely ground, insoluble minerals, either natural materials
or by-products, which immediately affect the rheological behavior of the fresh
concrete, but whose chemical effects take several days or months to manifest
themselves. A summary of commonly used concrete admixtures, and the appli-
cable ASTM specifications, is given in Table 8.
Many admixture operations are dependent upon the conditions under which
the admixtures ars applied* For instance, trlethanolamlne has been shown to
retard the hydration of C.S but to accelerate the hydration of C^A
(Ramachandran 1976a). Lignosulfonate acts as a set retarder and, in the
proper dosages, also increases strength (Young 1972). Admixture dosage is
critical if the desired effects are to be attained. Parameters such as water
content of the concrete formulation also play an Important role. For example,
amines and amino acids show accelerating properties at low water content, but
become retarders if the water content is sufficiently increased.
SURFACE-ACTIVE CHEMICALS (SURFACTANTS)
Surface active molecules are those which lower the surface tension of
water; examples are soaps and detergents. See Camp et al. (1985) for a
general discussion of surfactants. Two major uses of surfactants as concrete
22
-------
TABLE 8. COMMONLY USED CONCRETE ADMIXTURES
Primary function
Principal active ingredients/
ASTM specifications
Side
effects
Water-reducing
Normal
High rang*
Salts, modifications and derivatives
of lignosulfonic acid, hydroxylated
carboxylic acids, and polyhydroxy
compounds. ASTM C 494 (Type A).
Sulfonated naphthalene or melamine
formaldehyde condensates. ASTM C
494 (Type F).
Lignosulfonates may
cause air entrain-
ment and strength
loss; Type A
admixes tend to be
set-retarding when
used in high
dosage.
Early slump loss;
difficulty in
controlling void
spacing when air
entrainmenc is
also required.
Set-controlling
Accelerating
Retarding
Calcium chloride, calcium formate,
and triethanolamine. ASTM C 494
(Type C).
Same as in ASTM Type A; compounds
such as phosphates may be present.
ASTM C 494 (Type B).
Accelerators con-
taining chloride
increase the risk
of corrosion of
the embedded
metals.
Water-reducing and set-controlling
Water-reducing
& retarding
Sao* at used for normal water
reduction. ASTM C 494 (Type 0).
Water-reducing Mixtures of Types A and C. ASTM C
& accelerating 494 (Type E).
High-rang* Sam* as used for Type F with ligno-
vater-reducing sulfonatea added. ASTM C 494
4 retarding (Typ* G).
(See Typ* A above.)
(See Typ* C above.)
(See Typ* F above.)
(Continued)
23
-------
TABLE 8. (Concluded)
Primary function
Principal active ingredients/
ASTM specifications
Side
effects
Workability-improving
Increasing
consistency
Reducing
Segregation
Water-reducing agents, [e.g.,
ASTM C 494 (Type A)].
(a) Finely divided minerals (e.g.,
ASTM C 618).
(b) Air-entrainment surfactants
(ASTM C 260).
(See Type A above.)
Loss of early
strength when used
as cement replace-
ment .
Loss of strength.
Strength-increasing
Sao* as listed under ASTM C 494
(Types A, D, F, and G).
By water-
reducing
admixtures
By pozzolanic & Sane as listed under ASTM C 618
cementitious and C 989.
(See Types A and F
above.)
Workability and
durability may be
improved.
Durability-improving
Frost action
(air entrain-
ing)
Thermal- crack-
ing
Alkali-
aggregate
expansion.
Acidic
solution*
Sulfate
solution*
Wood resins, proteinaceous materials,
and synthetic detergents (ASTM C 260)
Fly ashes and raw or calcined natu-
ral porzolans (ASTM C 618); granu-
lated and ground iron blast-furnace
slag (ASTM C 989); condsnsed silica
fume; rice husk ash produced by con-
trolled combustion. (High-calcium
and high-alumina fly ashes, and slag-
Portland cement mixtures containing
less Chan 601 slag may not be sulfate
resistant.)
Strength lose.
Loss of strength at
early ages, except
when highly pozzo-
lanic admixtures
are used in con-
junction with
water-reducing
agents.
Adapted from Mehta (1987).
24
-------
admixtures art for reduction of the proportion of water in concretes (water-
reducing admixtures) and for introducing large numbers of small air bubbles
into the concrete mix (air-entrainment admixtures).
The materials that have been described and are generally available for
use as water-reducing admixtures and set-controlling admixtures fall into five
general classes:
1. Lignosulfonic acids and their salts.
2. Modifications and derivatives of Lignosulfonic acids and their salts.
3. Hydroxylatad carboxylic acids and their salts.
4. Modifications and derivatives of hydroxylated carboxylic acids and
their salts.
5. Other materials, which include:
(a) Inorganic materials, such as zinc salts, borates, phosphates,
chlorides.
(b) Amines and their derivatives.
(c) Carbohydrates, polysaccharides, and sugar acids.
(d) Certain polymeric compounds, such as cellulose ethers, melamine
derivatives, naphthalene derivatives, silicones, and sulfonated
hydrocarbons.
These admixtures can be used either alone or in combination with other organic
or inorganic material*.
Water-Reducing Surfactants
When water is added to cement, the mixture does not become veil dispersed
because of the high surface tension of the water and the tendency of the
cement particles to adhere and form floes due to the attraction of oppositely
charged edges of the cement particles. Water-reducing surfactants, or plasti-
cizing surfactants, act by adsorbing to the surface charges of the cement par-
ticles so that the surface of the particle becomes hydrophilic and no longer
attracted to other cement particles. Typical examples are citric acid,
gluconic acid, and lignosulfonlc acid. A better dispersed suspension of the
cement paste means that 1) at a given water-cement ratio, the cement paste has
a higher consistency and pumpability, 2) a lower water/cement ratio can be
used to increase strength and lower permeability without a change in consis-
tency, and 3) less cemant can b« used in the mix at the same water-cement
ratio (more aggregate or vast* loading).
Superplasticizers, sometimes called high-range water-reducing admixtures,
are 3 to 4 times as effective as normal water-reducing admixtures (Malhotra
1979). They are long carbon-chain, high molecular weight anionic surfactants
with a large number of polar groups. Adsorbed onto the cement particles,
superplasticizers import a strong negative, charge to their surface which helps
lower the surface tension of surrounding water and greatly increases the flu-
idity of the system. The better dispersion of the cement particles does tend
to cause their more rapid hydratlon, increasing the rats of set. Set retar-
dants are sometimes addsd to offset this effect.
. Water reducing admixtures may be useful in waste S/S with cement. The
5- to 10-percent water reduction made possible with normal plasticizers, or
25
-------
1
the 25- Co 30-percent water reduction with superplasticizers, increases the
compressive strength and decreases the amount of void space and therefore the
permeability of the final product for a given amount of cement. A more leach
resistant and durable S/S waste product should result.
Air-Entraining Surfactants
These materials differ from the water-reducing surfactants in that the
molecule contains a nonpolar portion along with the anionic, polar group. The
surfactants coae the surface of small air bubbles and stabilize them so that
they do not coalesce and separate from the concrete during mixing or setting.
They also coat the charged surfaces of the cement particles, causing them to
become hydrophobic, which can cause a lower rate of hydratlon (delayed set-
ting) at higher doses. Doses of 0.05 percent by weight of cement are often
sufficient to incorporate 0.05- to I-mm bubbles into the cement. Air contents
of 3 to 8 percent of the total mix are recommended for the production of
frost-resistant concretes, depending upon the size of aggregate* (larger
aggregates require lest entrained air) and the degree of exposure (more severe
exposure requires more air entrainment).
SET-CONTROLLING CHEMICALS
The early events in the setting and hardening of concrete involve the
dissolution, lonization, and hydration of sparingly soluble chemical compo-
nents. Any added soluble chemical that interferes with any of these processes
may delay or accelerate the rate* of hydration of the various cement compo-
nents, depending on its relative effect on all of the components present.
Hydratlng cement mixtures are composed primarily of acidic anions (sili-
cate and aluminate) and a base cation (calcium). The. solubilities of each of
these ions is dependent upon the concentrations of the other ions in solution.
Sinca most chemical admixtures which affect the rate of hydration of cement
pastes ionize in water solution, their presence is believed to alter the type
and concentrations of the ionic constituents in the solution phase, which
influences the dissolution and crystallization of the cement compounds. The
following guidelines are proposed by Joissl (see Mehta 1987):
1. Accelerators muse promote the dissolution of the calcium cations and
anions from the cement particles. The most effective accelerator
would promote the dissolution of ths least soluble constituent
(silicate).
2. Retarding admixtures must interfere with the dissolution cations
(calcium) and anions, preferably that anion which has the highest
rate of dissolution during aarly hydration (i.s., the eliminate
ions).
3. A strong cation (e.g., K+ or Na+) will reduce ths solubility of the
calcium cation but will accelerate the solubility of ths anions
(silicate and aluminate). At lower concentrations the former effect
is dominant; at the higher concentrations, the latter.
26
-------
4. A strong anion (e.g., Cl", NOT, or SO^) reduces the solubility of che
cement anions (silicates and aluminates) but tends to increase the
mobility of the calcium cation. At low concentrations the former
effect appears dominant; at higher concentrations, the latter.
From these considerations, it can be seen that the net effect of any
admixture on the rate of hydration depends on the balance of its effect on
the type and concentration of ions contributed by the admixture. For
instance, small concentrations of the salt of a weak base and a strong acid
(e.g., CaCl,), or a strong base and a weak acid (e.g., K2CO.), will tend to
retard the solubility of the calcium and aluminate ions. At higher concentrar
tlons, the accelerating effects of the increased solubilities of silicate and
aluminate ions predominate, and the salts become accelerators.
Calcium chloride at 1 to 3 percent the weight of cement is the most com-
mon accelerator and has been widely studied (Ramachandran 1976b, 1984). Some
of the properties of hydrating cement paste which are influenced by calcium
chloride are summarized in Table 9. Other salts that accelerate cement paste
hydration are sodium fluoride, aluminum chloride, sodium aluminate, and potas-
sium carbonate. Setting tim« as short as 15 to 30 seconds can be obtained.
Some ready-to-use mixtures have an initial set of 1 to 4 minutes with a final
sec of 3 to 10 minutes.
Organic acids and their weakly acid soluble salts act at accelerators
because they Increaae the dissolution of calcium. Calcium formate and formic
acid are common accelerators. Triethanolamine it an active accelerator
because of its ability to accelerate the hydration of C.A and the formation of
ettringite; however, it also retards the hydration of C^S and reduces the race
of strength development. Comparison of the development of strength and of
setting times- for aome common accelerators is illustrated in Table 10.
Set retarders also work by other mechanisms. Surfactants such as gluco-
nates and lignosulfonates delay bond formation among the hydration products,
delaying crystallization (Hansen 1959, Young 1976). Other retarders such as
sodium salts of phosphoric, boric, oxalic, and hydrofluoric acid form
insoluble calcium salts which precipitate out and form insoluble and imperme-
able barriers around th« cemsnt particles. This slows hydration considerably
and retards see and strength development.
Early strengths ars lower in specimens with retarders than those in ref-
erence specimens, as would-be expected. However, ac longer periods the speci-
mens containlnf retarders generally have higher compressive strength and
comparable flexur.1 strength (see Tables 11 and 12). The drying-shrinkages of
specimens containlnf retarders are also comparable to those reference, speci-
mens without admixtures. Soluble calcium salts which may provide anions that
adsorb onto the calcium hydroxide crystal surfaces cause a retarding effect
(e.g., calcium nitrate).
Scholer (1975) studied the effects of 65 different chemical compounds and
three ASTM Type D admixtures on the fresh and hydrated condition °* «»«
pastes. Setting time, molecular configuration related to the effectiveness of
?he retarders, oven-dry shrinkage, nonevaporable water, and specific «»'"
area were determined for each admixture. Long-term shrinkage of concrete was
27
-------
TABLE 9. SOME OF THE PROPERTIES INFLUENCED BY THE USE
OF CALCIUM CHLORIDE ADMIXTURE IN CONCRETE
Property
General effect
Remarks
Setting
Compressive
strength
Tensile
strength
Flerural
strength
Heat of
hydration
Resistance to
sulfate attack
Alkali-aggregate
reaction
Corrosion
Reduces both Initial and
final setting
Increases significantly
the compressive strength
in the first 3 days of
curing (gain may be
about 30-100Z).
A slight decrease at
28 days.
- "
'A decrease of about IOZ
at 7 days.
An increase of about 30Z
in 24 hours.
Reduced
Aggravated
Causes no problems in
normal reinforced con-
crete if adequatt pre-
cautions taken. Dosage
should not exceed 1.5Z
CaCl.i and adequate cover
to be given. Should not
be used in concrete con-
taining a combination of
dissimilar metals or
where thert is a possi-
bility of stray currents.
(Continued)
ASTM standard requires Chat
initial and final setting
times should occur at least
1 hour earlier with respect
to reference concrete.
ASTM requires an increase of
at least 125Z over control
concrete at 3 days. At
6-12 months, requirement Is
only 90Z of cpntrol specimen.
This figure may vary depending
on the starting materials and
method of curing. The
decrease may be more at
28 days.
Total amount of heat at longer
times is almost the same as
that evolved by reference
concrete.
Can be overcome by use of
Type V cement with adequate
air entrainment.
Can be controlled by use of
low-alkali cement or
potzolana.
Calcium chloride admixture
should not be used in pre-
stressed concrete or in a
concrete containing a com-
bination of dissimilar
metals. Some specifications
do not allow use of CaCl. in
reinforced concretes.
28
-------
1
TABLE 9. (Concluded)
Property
General effect
Remarks
Shrinkage and
creep
Volume change
Resistance to
damage by freez-
ing and thawing
Watertightness
Modulus of
elasticity
Bleeding
Increased
Increase of 0-15Z reported.
Early resistance improved.
Improved at'early ages.
Increased at early ages.
Reduced
At later ages may be leas
resistant to freezing and
chawing.
At longer periods almost same
with respect to reference
concrete.
Adapted from Mehta 1986, p 263.
TABLE 10. EFFECT OF ACCELERATORS TRIETHANOLAMINE, CALCIUM FORMATE,
AND CALCIUM CHLORIDE ON THE RATE OF HARDENING
AND COMPRESSIVE STRENGTH OF CONCRETE
Admixture
None
Triethanolamin*
None
CaCl-
Calcium format*
Calcium format*
Calcium format*
Calcium format*
Rat* of us*
(w* X
of cement)
^
0.025
M
2.00
O.OS
1.00
1.50
2.00
Comprtssiv*
strength (psi)*
7-day
4,810
5,160
4,430
4,490
4,630
4,910
5,200
5,330
28-day
6,710
6,670
6,130
5,615
6,275
6,485
6,585
6,610
Tim* of final
setting (hr)
12.0
9.7
12.4
4.0
8.9
8.4
7.6
6.7
Source: Rosskopf, Linton, and Peppier 1975.
* To convert pounds (fore*) per square inch to megapascals, multiply by
0.006894757. '
29
-------
TABLE 11. COMPRESSIVE STRENGTH (CS) OF MORTARS CONTAINING RETARDERS
Admixture 1 day
Retarder (wt I of cement) CS
None
Sucrose
Sucrose
Glucose
Glucose
Phosphoric acid
Phosphoric acid
Phosphoric acid
««»
0.5
1.0
1.0
2.0
0.5
1.0
2.0
11. 8
10.0
1.3
7.1
1.0
7.1
2.2
1.2
2 days
CS
21.6
21.6
11.8
23.7
8.3
18.1
14.7
12.3
7 days
CS
37.8
47.1
43.2
36.8
27.9
48.1
45.1
44.1
28 days
CS
45,3
59.8
53.9
53.4
45.6
60.8
64.7
60.3
90 days
CS
53.9
62.3
60.3
58.3
51.5
71.1
74.0
69.6
Note: All values given in megapascals (1 MPa - 145 pai)
TABLE 12. FLEXURAL STRENGTH (FS) OF MORTARS CONTAINING RETARDERS
Retarder
(wt
Admixture
Z of cement)
1 day
FS
2
days
FS
7
days
FS
28
days
FS
90
days
FS
Nona 3.5 4.8 7.6 8.6 8.8
Sucro«e 0.5 2.9 5.0 7.8 8.1 8.2
Sucrose 1.0 0.4 2.8 7.6 7.9 9.4
Glucose 1.0 2.0 4.9 6.7 7.5 7.9
Glucose 2.0 0.1 2.5 5.5 7.4 7.9
Phosphoric acid 0.5 1.8 4.2 7.6 8.3 8.'l
Phosphoric acid 1.0 0.5 3.3 7.6 8.5 8.6
Phosphoric acid 2.0 0.2 3.2 7.7 7.5 8.0
Note: All values given in megapascals (1 MPa 145 psi).
not appreciably affected. He concludtd that strong retardtrs have a molecular
composition that includes many oxygen atoms that are constrained to approach
each other closely. Hydroxyl, carboxyl, and carbonyl arc all affective, but
carbonyl is especially strong in its influence on set time. These groups are
presumed to exert a polarizing influence that contributes to the strong
adsorption on the) solid surfaces. More veakly electronegative atoms on the
molecule do noe have the same effect as oxygens.
The retarding effects of sugars on cement hydration were studied in
detail by Thomas and Birchall (1983). They compared different sugars using
solution analysis, colorimetry, calcium binding ability, and alkaline stabil-
ity. The best retarders were sucrose and raffinose, which also have a remark-
able ability to solubilize cement constituents and give rise to dramatic
increases (10 to 20 times) in the amount of silica present in solution.
1
ever, no evidence for sucrose-silicate complexes was found using C and "'Si
nuclear magnetic resonance (NMR) analysis. The authors relate the retarding
action of sugars in terms of adsorption onto and the poisoning of hydrating
surfaces.
30
-------
MACRO-DEFECTfFREE (MDF) CEMENT PASTES
A new class of high-strength cement materials, called macro-defect-free
(MDF) cements, achieves higher strengths by the absence of the relatively
large voids or defects which are usually present in conventionally mixed
cement pastes (Roy 1987). In this process, 4 to 7 percent of one of several
water-soluble polymers, such as hydroxypropylmethyl cellulose, polyaerylamide,
or hydrolyzed polyvinylacetate, is added as a rheological aid to permit cement
to be mixed with very low water-cement ratios. Coupled with high-shear
mixing, tha method produces a plastic, cohesive mixture that can be shaped by
extrusion or other forming techniques. Set times range from minutes to hours.
The polymers also appear to be a significant structural component. Where
ordinary cement mixtures have compressive strengths of around 40 MPa (5,800
psi), MDF cements can have a compressive strength as high as 100 to 300 MPa
(14,500 to 43,500 psi).
MINERAL ADMIXTURES
Mineral admixtures are usually siliceous materials that are added to the
cement mixtures in relatively high proportions, for example 15 to 100 percent
of the weight of cement. Typical admixtures are pozzolanic (e.g., volcanic
glass, silica fume, rice husk ash, and Class F fly ash), cementitious (granu-
lated iron blast-furnace slag, hydraulic liaes, and slag cements), or both
(high-calcium Class C fly ash). Other non-setting mineral admixtures are
finely divided quartz; siliceous sands; dolomitic and calcitlc limestones;
marble, granite, and other rock dusts; and hydrated dolomitic or high-calcium
lima. Of these, fly ash, kiln dusts, and blast-furnace slags are by far the
most commonly used mineral admixtures*
Pozzolanic Admixtures
Proposed minimum standards for fly ash used as a cement admixture are
given by Idorn and Thaulow (1985). For Class F fly ashes, tha major constitu-
ent, SiO., should comprise a minimum of 50 percent of tha material by weight;
Al.O- should comprise 20 to 35 percent; and CaO, 2 to 8 percent. Between 60
and 85 percent of tha volume of tha fly ash should ba noncrystalline or glass.
Pozzolanic materials, by definition, react with hydratad calcium hydroxida to
form C-S-H,
Pozzolan + CH + H.O »2(
while la Portland cement the raacelon is
fa
-------
resulting from pozzolan addition is thought to be the reason for lower leach-
ing losses, although a certain amount of additional binding of heavy metals by'
the active silica (thus preventing their leaching) may be responsible for some
of the effect (Coti and Webster 1987, Van der Sloot et al. 1987).
The setting and strength characteristics of Portland cement products con-
taining Class C (high lime) fly ash were studied by Baker and Laguros (1985).
Class C fly ash from western coal has not been studied to such an extent as (
Class F low-lime, eastern fly ash. They found that setting time and develop-
ment of compresslve strength were delayed, but after 1 week, the compressive
strength of all mixtures containing Class C fly ash (at 20 to 50 percent
replacement of cement) surpassed the cement-only controls. A decrease in the .
level of calcium hydroxide was not seen (as is typical for mixes with Class F
fly ash). They suggested that the retardation mechanism may be associated
with the high levels of ettringite formed early in the hydracion process and
its subsequent conversion to monaulfoaluminate. A simple heat of hydration
test was presented to help explain the observed strength gains. High- and
low-calcium fly ash-cement mixtures were also compared by Grutzeck et al.
(1985), who found similar results, with high-calcium mixtures retarding early
strength development to a larger extent.
The primary commercial advantage of mineral admixtures is to lower the
amount of cement required to attain a given strength in the final structure.
Adding a mineral admixture Co a given cement mixture produces multiple
effects: improved workability, retardation of see and reduced thermal crack-
ing, increased ultimate strength, increased impermeability, and better dura-
bility to chemical attack such as by sulfate (Berry and Maihoera 1982).
The use of mineral additives has been studied as a method of reducing the
general permeability of cured cement matrices containing low-level nuclear
waste (Carlson 1987). Improved admixtures have exhibited markedly superior
leach resistance, even for very mobile, radioactive cesium.
Soluble Silicate Admixtures
Soluble silicates have been added to waste S/S system* and were a part of
an early patented process (Jones and Malone 1982). Recent studies (Davis et
al. 1987) have confirmed the benefit of soluble silicate addition. These
studies showed that added silicate lavsls up co 5 percent of ths weight of the
waste resulted in a 25- to 90-percent reduction in the metal ion concentration
of an acid (pB 5) Itachate. In S/S waste products including aggregates, the
larger the aggregate, ths larger the effect of added silicate.
Effects of soluble silicates have traditionally been credited to the for-
mation of metal silicates that are more stable and less soluble than metal
hydroxides (Falcone ct al. 1984). However, data of Davis et al. (1987) using
a simple penetration model suggested that ths primary influence of ths soluble
silicate is due to a reduction in permeability and improved waste integrity.
Ths soluble silicats apparently reacts with ths fres lias producsd in the nor-
mal cement hydration process, decreasing the porosity and thus the perms-
ability in a manner similar to the effects of added reactive silicates in fly
ash or blast furnace slags, as discussed above.
32
-------
POSSIBLE MECHANISMS OF THE EFFECTS OF
ORGANIC COMPOUNDS ON PORTLAND CEMENT PRODU6K1
. /
The variables associated with che use or admixtures in concrete produc-
tion are numerous. These variables are magnified when a toxic waste is being
considered for S/S. The variable and complex composition of a typical hazard-
ous waste increases the complexity of evaluating the S/S process. The poten-
tial mechanisms for chemical Influence of organic materials on Portland cement
muse be studied and understood. However, even then, the prediction of che
effects of specific levels of individual waste constituents on final product
characteristics will be difficult at best (Hans en 1959).
An example of the complexity surrounding the interaction of organics wich
cement matrices has been reported by Cartledge et al. (1987). Several phenols
with different substituents on the aromatic ring were solidified with types I
and III Portland cement at several concentrations up to 20 percent by weight
of cement. At various times of cure, the products were extensively investi-
gated using water and solvent extraction* optical and scanning electron micro-
scopy, X-ray diffraction, and NMR spectroscopy. Although all phenol
derivatives were readily leachad from all specimens, significant differences
in physical characteristics (such as setting times and compressive strengths) ,
the morphology of the cement matrix, and the nature of the crystalline phases
were observed. The authors stata that even subtle differences in the struc-
ture of che organic may result in major differences in eha nature of che
cement-organic interaction.
Several conceptual models of tha interfarenca mechanism* of organics on
cement itious and pozzolanic are addressed below. These potential mechanisms
include interference via adsorption, complexation, precipitation, and nuclea-
tion (Young 1972, 1976).
Adsorption
'One possible intarfaring mechanism la adsorption of added molecules onto
tha surface of tha cement particles, chus blocking tha normal hydratlon reac-
tions. Studies by Young (1972) have shown that tha retarding effect of
organic compound* is related to tha strength of chair adsorption on mata-
s table, hexagonal* calcium aluainata hydrates. In this way tha organic com-
pound is thought to inhibit crystal grovth and conversion to C.AHg* Th*
inhibiting effaet roughly correlates with tha number of hydros?! "e.g.,
alcohols, glycerol, or sugars), carboxylic (a.g., acids, acetic acid, citric
acid), and carbonyl (a.g., ke tones, dimethyl katona, mathyl athyl katona)
groups in tha organic molecule. Hanssn (1952, 1959) noted tha effect of two
particular famlliaa of organic compounds (lignosulfonic acid derivatives and
hydroxylatad carboxylic acids) on satting reaction*. Lignosulfonataa ara
strongly adsorbed onto C.A (Kawoda and Niahiyam* I960. Blank et al. 1963).
Tha adsorption of calcium lignowlfonat* onto C.A raaulti ia a relatively
thick film or layer, which is indicative of a cnemicai reaction involving the
organic and C.A hydration production. Adsorption of tha hydroxylatad ^
carboxylic acids, such as gluconata or oxalates. involvaa simultaneous Ca
ion adsorption to tha carboxyl groupa and adsorption of silicata iona by the
hydroxyl groups (Figure 3a). Taplin (1962) found retarding affacta from
33
-------
o
N
0
s
f
\
R
\
)
0'-
0 R
>\ / 0 f
\ /' 9
..... \Z
\
\
)
(a) (b) (c)
CRYSTAL SURFACE OF ANHYDROUS COMPOUNDS
Figure 3. Possible adsorption mechanism* of organic admixtures onto
surfaces. (Note: C, H, 0, M, R carbon* hydrogen and oxygen
atoms, metal, ions, and rest of the molecule, respectively.)
(Source: Hansen 1952.)
aliphatic and aromatic dicarboxyllc acids (e.g., maleic acid). In alkaline
solutions where maleic acid has no hydroxyl group for hydrogen bond adsorp-
tion, chelation may be the mechanism of see retardation (Figures 3b and 3c).
The overall retardation of hydratlon can be the result of an initial
acceleration followed by retardation. This effect has been observed with C.A
in the presence of sugars and also for C-S hydratlon in the presence of ligno-
sulfonates (Seligmann and Greening 1964, Chattarji 1967, Daugherty and
Kowalewski 1968). The organic compound, when adsorbed onto Che cement
clinker, first assists in hydration, and later retards hydration by adsorption
onto the hydration product. These effects are altered by the presence of sul-
fate ions, which Increases the formation of ettringite and lovers that of
C.AHg because the adsorption of calcium llgnosulfonate or sugar onto C.A is
reduced. The morphology of ettringite can also be modified by the presence of
sugars (Daugherty and Kowaltwski 1968).
The adsorption of organic retarders by C.S and C.S in an aqueous solution
la much weaker than adsorption by C.A. However, the amount adsorbed (3 to
5 mg/g on C.S and 1 to 2 mg/g on C.S) is sufficient to form a multilayer film
on the C.S and C.S. The occurrence of calcium ion concentrations in the
aqueous solution (Figure 4) soon after the addition of retarders indicates
chat the retarding reaction is not with C.S or C.S, but rather with other
hydration product*. Strong retarders can extend indefinitely the hydration of
C.S. Addition of hydrated C.S or calcium hydroxide can curtail the retarder
effect and permit hydration.
Although data indicate the stronger adsorption of organic compounds by
C.S, che organic retarders act primarily by retarding the C.S hydration.
Therefore, most of the retarder is adsorbed by C.A, but sufficient amounts
must be present to be adsorbed by C.S to actually retard Che set. In some
instances, differential adsorptive properties and hydration dynamics make it
advantageous to delay addition of the retarding agent (Bruere 1963, 1966).
When addition of the retarder Is delayed, che adsorption of C.A is reduced,
and adsorption on C.S in increased. The initial set is not hindered; however,
the retardation effect on the final set is increased.
34
-------
LEGEND
NO ADMIXTURE
SUCROSE
TORTONIC ACID
. TIME. HOURS
Figure 4. Influence of some organic admixtures
on soluble CaO and SiO- during «arly
hydracion of trlcalciua silicaca.
(1 we X additiona) (Source: Young 1972.)
Although adsorption of organic retardera ia primarily on C.A, retardation
is dua to adsorption on C.S. Thara is no evidence of adsorption on C-S. Nor
is thtre avidanca of adsorption onto anhydrous surfacaa. Organic addleivaa
can hava an important baaring on raaction ratat during camant hydracion. Hov-
evar, adsorption ia not tha only concaptual modal that can account for aat
retardation.
Complaxation
Taplin (1962) ralatad tha ratarding activity of tha organic compounda ha
studied to tha proximity of oxygen atoma to carbon atom*. Ha observed thac
compounds with oxy-functlonal group* in proximity to each other vara more
effective as ratardara. Ha theorized that chelation to metal iona waa an
important factor in sat retardation. Calcium ions can chalata with various
hydroxyl or carboxylic acids, but tha retardar or accelerators (raapectively)
are so dilute that complexation of the calcium could not be an important fac-
tor (Young 1972).
The effects of complexing calcium are more significant when the additive
to affected-ion ratio ia large, and when the effected ion ia important to the
setting system. Such would be the case for tha aluminata and farrite ions.
35
-------
1
Several workers (Kalousek tc al. 1943, Suzuki and Nlshi 1959, Roberts 1967)
have shown chac che addition of sucrose increases the concentration of alumina
and calcium ions to above-normal levels. Experiments (Young 1972) with C-A
indicate that 1 vt I additions of sucrose, succinic acid, or tartaric acid
increase che amounts of calcium and alumina in solution at first, but concen-
trations later decrease to normal or below normal. Silica concentrations are
also increased when additives that affect alumina concentrations are used.
Apparently, conditions la a cement paste are favorable for aluminate, ferrite,
and silicate ion complexation. It is possible that complexatlon delays the
formation of hydration products. When cement crystal-forming ions are kept in
solution by complexation, hydration barriers are established that retard the
set.
Precipitation
The formation of insoluble hydration products by additives reacting with
cement compounds has been reported conceptually not to be a realistic mecha-
nism of admixture interference (Young 1972). Certainly, the formation of
insoluble compounds could impede water transport, solubility, and subsequent
hydration reactions. However, if retardation is due to precipitation, then
the process should be nonselective. In this cast, C.S and C.A should both
release calcium ions, and the resulting effect on setting should be equally
weighted in regard to C.S and C.A activity. Regardless, the retardation is
known to be related primarily tS CjA content (Young 1972).
However, if the precipitation reaction involves the hydration product
itself, the precipitated product would preferentially ba produced at the sur-
face of the hydratlng material. This is thought to be the mechanism of action
of phosphates, boratas, and oxalates (Mehta 1987).
Nucleation
The inhibition of nucleation of crystalline calcium hydroxide by soluble
silica, which is present in small quantity, is believed to be the self-
retarding set feature of C.S hydration. Growth of a crystalline matrix is
retarded by the adsorbed silica ions when * C-S-H layer results in a
diffusion barrier to calcium hydroxide. Eventually, crystal growth results in
the adsorbed silica being trapped in the crystalline matrix as the hydration
process continues. Prismatic growth of calcium hydroxida results from differ-
ential adsorption of silica on calcium hydroxide crystal faces (Young 1972).
It is postulated that organic retarders act much the sama as silica ions
in being adsorbed onto the calcium hydroxide nuclei. However, as the name
implies, orgsnlc rstardsrs ars much mora effective in being adsorbed and more
completely cover tha crystal growth surfaces. This results in part from more
retarders being solubllizsd. Tha resulting retardation of crystal gr°*tj
resulta in more crystallite nuclei forming in the saturated solution. The net
effect of -rystal growth on these many nuclei is responsible for the acceler-
ation of hydration following tha retardation period (Young 1976). With
about I ; at of a strong retarding agent, C.S hydration is completely
Inhibited. iditions of prehydrated C.S can overcome tha effect of the
organic rec^ler, lending support to afi adaorption/nucleation model.
36
-------
1
Several workers (Kalousek tc al. 1943, Suzuki and Nlshi 1959, Roberts 1967)
have shown chac che addition of sucrose increases the concentration of alumina
and calcium ions to above-normal levels. Experiments (Young 1972) with C-A
indicate that 1 vt I additions of sucrose, succinic acid, or tartaric acid
increase che amounts of calcium and alumina in solution at first, but concen-
trations later decrease to normal or below normal. Silica concentrations are
also increased when additives that affect alumina concentrations are used.
Apparently, conditions la a cement paste are favorable for aluminate, ferrite,
and silicate ion complexation. It is possible that complexatlon delays the
formation of hydration products. When cement crystal-forming ions are kept in
solution by complexation, hydration barriers are established that retard the
set.
Precipitation
The formation of insoluble hydration products by additives reacting with
cement compounds has been reported conceptually not to be a realistic mecha-
nism of admixture interference (Young 1972). Certainly, the formation of
insoluble compounds could impede water transport, solubility, and subsequent
hydration reactions. However, if retardation is due to precipitation, then
the process should be nonselective. In this cast, C.S and C.A should both
release calcium ions, and the resulting effect on setting should be equally
weighted in regard to C.S and C.A activity. Regardless, the retardation is
known to be related primarily tS CjA content (Young 1972).
However, if the precipitation reaction involves the hydration product
itself, the precipitated product would preferentially ba produced at the sur-
face of the hydratlng material. This is thought to be the mechanism of action
of phosphates, boratas, and oxalates (Mehta 1987).
Nucleation
The inhibition of nucleation of crystalline calcium hydroxide by soluble
silica, which is present in small quantity, is believed to be the self-
retarding set feature of C.S hydration. Growth of a crystalline matrix is
retarded by the adsorbed silica ions when * C-S-H layer results in a
diffusion barrier to calcium hydroxide. Eventually, crystal growth results in
the adsorbed silica being trapped in the crystalline matrix as the hydration
process continues. Prismatic growth of calcium hydroxida results from differ-
ential adsorption of silica on calcium hydroxide crystal faces (Young 1972).
It is postulated that organic retarders act much the sama as silica ions
in being adsorbed onto the calcium hydroxide nuclei. However, as the name
implies, orgsnlc rstardsrs ars much mora effective in being adsorbed and more
completely cover tha crystal growth surfaces. This results in part from more
retarders being solubllizsd. Tha resulting retardation of crystal gr°*tj
resulta in more crystallite nuclei forming in the saturated solution. The net
effect of -rystal growth on these many nuclei is responsible for the acceler-
ation of hydration following tha retardation period (Young 1976). With
about I ; at of a strong retarding agent, C.S hydration is completely
Inhibited. iditions of prehydrated C.S can overcome tha effect of the
organic rec^ler, lending support to afi adaorption/nucleation model.
36
-------
SECTION 5
EFFECTS OF WASTES ON CEMENT/POZZOLAN PROCESSES
INTRODUCTION
Acids, salts, bases, and organic materials may be present in hazardous
wastes singly and in combinations. As such, predicting their single and col-
lective effect on the durability and containment of the typical cement and
po_zzolan S/S processes presents a difficult problem for process designers and
regulatory agencies. Long-term effects are especially difficult to estimate
because subtle differences in environmental parameters can have significant
long-term consequences on the integrity of the S/S waste.
Early concerns about the effects of Industrial wastes in the environment
on concrete corrosion are described by Blczok (1967). .He lists industrial
wastes that are harmless to concrete and mortar products:
1. Brines containing bases but no sulfates.
2. Potassium permanganate, occurring at fermenting and purification
installations.
3. Sodium carbonate (soda) and potassium carbonate (potash).
4. Bases (caustic lyes of potash and soda, lime and ammonia), provided
their concentration is not excessively high.
5. Oxalic acid occurring at tanneries.
6. Mineral oils and petroleum products (benzene, kerosene, cut-back
oil, naphtha, paraffin, tar), as long as these contain no acids that
can continue to remain in the products after chemical treatment.
Industrial wastes deemed to be aggressive and detrimental to concrete and
mortar products are
1. Water containing gypsum, e.g., that used for quenching coal slag.
2. Ammonia salt*.
3. Hydrochloric acid, nitric acid, and aulfuric acid.
4. Chlorine and bromine.
5. Acetic acid.
6. Pur« alcohol (in certain cases only, e.g., absolute alcohol, owing to
the dehydrating effect).
7. All sulfur and magnesium salts.
8. Hydrogen sulfide and sulfur dioxide gas.
9. Animal fats.
10. Salts of strong acids, formed with magnesium, zinc, copper, iron, ,
aluminum, and other metals, the hydroxides of which are poorly
soluble in water.
11. Vegetable, animal, and some acidic.
Biczok (1967) also suggests that besides the wide variety of salts listed
above, the salts formed by weak bases and strong acids, such as ZnS04
37
-------
Zn(NO ) . FeS04, Fed , Fe(NO ) , CuSO,, CuCl , Cu(NO,),, Al,(SO,),, Aid,,
and AI(fl03)3. may be RegardedJaS potentially IggressiveT Furthermore,
chlorine, bromine, and iodine are considered to be aggressive. If sugar,
molasses, spirit, ammoniac, salt, strong acids or animal fats are present,
expert opinion based on literature information, practical experience, or
previous analyses should be obtained.
ORGANIC WASTES
A few studies have been directed toward the effects of typical organic
waste components upon cement and pozzolaa S/S processes. However, most of
these studies did not include extensive testing of the resultant S/S waste
products. Strength and leaching tests are complicated by the fact that the
setting waste/binder matrix is a dynamic system whose properties will vary
Considerably over the first few weeks or months after they are produced. For
instance, many of the materials that retard the initial setting of cement and
pozzolan products will result in higher ultimata strength and containment
roperties of tha solid (see Roberts 1978 and Smith 1979). A further compil-
ation is that Interference effects can vary with cement- or pozzolan-to-watet
.atio, interactions with other waste components, and even with order, type,
and timing of mixing.
Portland Cement Mixture!
The mechanisms by which organic constituents interact with cement matri-
ces is important to predicting tha long-term effects of S/S waste technolo-
gies. Ideally, long-term stability could be assayed using short-tern testing
techniques. Several testing techniques such as X-ray diffraction and electron
and optical microscopy are necessary to assess tha effects of waste constitu-
ents on the complex chemistry of cement setting reactions (Tittlebaum et al.
1986). There is also aoma information that the chemlcaljnatrix of inorganic
binders can increase £h« solubility and reactivity Of organic waste components
remann 1982V.Some organic* have greater water solubility and reactivity
in a strong acid environment, while others art more reactive under strong
alkaline conditions. Phenols, for example, are transformed to more reactive
phenolates in basic pRs. The water solubilities of some compounds liks kepone
also vary with pH.
Clark et al. (1982a,b) studied selected cement media to determine inter-
ference and retention capacities of an oil and a synthetic organic liquid
mixture (sea Table 13). Media used la the study included Portland type I
cement, vermlculita plus Portland type I cement, Nutek 380-cemant process
(Nuclear Technology Corporation), Portland type I cement with sodium silicate
and an oil emulsifier, Delaware Custom Materials' (DCM) cement process, and
U.S. Gypsum's Envlrostona Process. All of the processes were applied to
stabilization of oil (Inland Vacuum Industries No. 19 Vacuum Pump Oil), but
only tha Nutek and Envirostona processes were applied to tha synthetic organic
liquid. The solidified waste forms were evaluated only In terms of being
free-standing, monolithic solids and having no free liquids following solid-
ification. Other preliminary testing was conducted only on selected
specimensimmersion testing (72 hours soak to check for loss of integrity),
vibratory shock testing, and flame testing. Approximately 100 subjectively
determined formulations were prepared and examined.
38
-------
TABLE 13. COMPOSITION OF SIMULATED ORGANIC LIQUID WASTE
USED BY CLARK ET AL. (1982)
Component
Mcthanol
Ethanol
Acetone
Isopropanol
Diethyl ether
Ithyl acetate
n-Hexane
Benzene
Toluene
Acetonitrile
1,2-Ethylene dlchlorlde
Chloroform
Volume
22
22
9
5
9
4
9
4
4
4
4
4
Component
formula
CHOH
C H OH
CH3COCH3
CH3CHOHCH3
C2H5OC2H5
C.H-COOCH,
25 3
CH.(CH2).CH-
C6H6
C H CW
^6*5 3
CH.CN
C.H.Cl,
242
CHC13
In general, loadings of the oil up to 30 percent of the volume of final
product yielded free-standing solids that passed the immersion test. However,
loadings of over about 10 percent of the organic liquid waste cause loss of
physical integrity and the formation of free liquid from the solidified
product. The 10 formulations judged by Clark et al. (1982) to have the
highest loadings and which gave satisfactory monoliths using their criteria
(given only as "reasonable physical integrity") are given in Table 14.
Telles et al. (1984) presented a review of available treatment processes
used in the S/S of organic wastes. Table IS summarizes the organic contami-
nants reported and the S/S materials covered in their review. Telles et al.
(1983) presented some unpublished results from studies Involving fixation of
synthetic sludges containing chromium, cadmium, lead, and nitrocellulose. The
fixatives were polysilicates and amlne-cured epoxides. Inorganic dry sludge
containing Pb, Cr, and Cd was treated with polysilicates and with polysili-
cates in combination with Portland cement and fly ash. An organic wet sludge
containing nitrocellulose was also treated with polysilicates. The fixing
agents were mixed in five compositions that Included variable amounts of the
following: sodium silicate, Portland cement, fly ash, CaCl., HNO., and NaOH.
The mixtures of the various compositions were allowed to cure from 3 to 4 days
and then were subjected to extraction procedure (EP) toxicity testing to
determine the effectiveness of the fixation mechanism in regard to leaching of
contaminants. .
39
-------
1
TABLE 15. ORGANIC CONTAMINANTS AND FIXATIVES USED IN EXPERIMENTAL STUDIES
Contaminants
Fixatives
PCBs
Kepone
Various chlorinated
aliphatic and aromatic
hydrocarbons
Phenol, tolue'ne
Ion-exchange resins
Various pesticides
Oils
Mixed solvents*
Nitrocellulose
Dloxin
Cement-based processes/silicates, polyesters,
limes, polymer-cements
Cement-based/silicates, sulfur, epoxide
Cement-based/proprietary organic polymers,
gypsum
Cement-based/silicates
Cement-based/silicates, polymer-cements,
epoxldes, polyester, asphalt, polyethylene
gypsum, ureas
Polymer-cements, polybutadiene/polyethylene
(surface encapsulation)
Cement-based/silicates, polymer cements,
proprietary polymers, ureas
Gypsum, polymer-cements* cement/silicates
Cement/silicate, tpoxide
Calcium oxide, asphalt, Portland cement,
asphalt/sulfur
Source: Telles et al. 1984.
* A mixture comprised of the following: alcohols, ketones, aldehyde*,
esters, ethers, alkanes, acetonitrite, and chlorinated cleaning solvents.
Acceptable leaching performance for calcium-polysilicate fixed contami-
nants was only obtained In these test* at or below the 0.2-percent weight
(2,000 ppm) loading of Pb. Mixtures with Cd, Cr, and Pb were also managed at
the 0.2-percent weight loading. Results of tests on contaminants fixed with
sodium silicate and Portland cement Indicated that samples with contaminant
loadings as high as 24,000 ppm (2.4 percent) were able to meet EPA toxlcity
criteria. Organic sludges containing nitrocellulose were not acceptably sta-
bilized using polysillcates and Portland cement.
Buchler (1987) studied the adsorption of water-soluble organic molecules
onto a tetramethyl ammonium chloride (TAG) derivative of Wyoming bentonite
clay as a possible S/S pretreatment method. The hydrophobia nature of the TAC
compared with the normal sodium cation makes the clay more receptive to
organic molecules, especially polar organics. The strength of the adsorption
was inversely proportional to the temperature so that higher leaching losses
would be expected in the summer months. Treatment of 1,000-ppm phenol wastes
at 20* C produced an effluent containing 150 ppm phenol; effectiveness
decreased at lower phenol concentrations and higher temperatures. Glucose,
glyclne, and dextran were also tested but showed a lower adsorption pattern.
41
-------
1
In a similar study, Sheriff et al. (1987) used activated charcoal as well
as tetra-alkylamnonium substituted bentonite clays as prestabilization adsor-
bents before incorporating them in a number of cementitious matrices. They
found a clear trend between the degree of polarity of substituted phenols and
their ability to be absorbed by the substituted clays. Activated charcoal
effectively adsorbed up to a 1,000-ppm concentration of the phenols and
chlorinated phenols, reachability and physical strength measurements are to
be carried out on the S/S products.
Newton (1987) described an organic S/S system in which a fast sorption
interaction occurs between a sodium magnesium fluorolithosilicate and/or
sodium bentonite that has been reacted with a quaternary ammonium compound.
Other admixtures are also added to decrease the permeability of the final
product and to improve mixing and leaching characteristics. The formulation
is being sold as a proprietary S/S agent that is claimed to have been
successful in treating a variety of organic-bearing wastes.
An amorphous silicate glass foam of 8 to 200 mesh spheres with numerous
cells has been suggested to be the universal sorbenc material (Temple et al.
1978). The material is free-flowing granules with a loose bulk density of
2 Ib/cu ft (32 kg/cu m). The material is said to be an excellent absorbent
for acid, alkalies, alcohols* aldehydes, arsenatas, ketones, petroleum
products, and chlorinated solvents. 1C absorbs more than 10 times as ouch
liquid as expanded clays, and its sorbenc capacity ranges from 0.77 gal/lb
(6.4 cu dm/kg) for perchloroethylene to 1.82 gal/lb (IS cu do/kg) for
phosphoric acid. The loaded material has safely been solidified using
Portland cement and sodium silicate, but no testing results were available.
Roberts (1978) and Smith (1979) report results of an extensive study of
the effects of selected organics upon the strength and leaching characteris-
tics of a typical lime/fly ash S/S formulation. Organic compounds Included
methanol, xylene, benzene, adipic acid, and an oil and grease mixture.
The effects of ethylene glycol on cement-based solidification processes
have been documented (Chalasani et al. 1986, Walsh tt al. 1986). These inves-
tigators used a combination of techniques Including extraction with solvents
of varying polarity, scanning electron microscopy, and X-ray powder diffrac-
tion. The ethylene glycol appears to occupy three different sites in the
solidified product as ascertained by their different extractability. No visi-
ble gross alteration in morphology of the cement matrix is apparent until the
ethylene glycol reaches around 10 percent of the cement by weight. Level*
above 10 percent clearly weaken the structure and cause higher leachability of
the compound. Some evidence from X-ray diffraction indicates that the eth-
ylene glycol moves into the 0.3- to 2.5-nm voids in the C-S-H gel phase and
alters its structure. A major effect of the organic addition is the retarda-
tion of setting reactions so that significant changes in the matrix occur for
up to at least 1 year of curing time. ^ _________'-
Conner (1984) describes the successful solidification of 4 million gal-
lons of waste containing 6.6 percent ethylene glycol, 2.5 percent sodium
terephthalate, 8.3 percent sodium chloride, and 234 mg/t antimony. The final
product had a UCS of 55 to 70 psi (0.38 to 0.48 MPa) and leached 0.1 m«,/l
-------
1
antimony, 5 mg/i sulfate, 160 og/Z chloride, and 350 tng/l chemical oxygen
demand. No details on the type of reagents or leaching tests used were given.
Gilliam et al. (1986) studied the immobilization of four waste streams in
simple cement grout. On* of the wastes contained 2,230 ppm PCBa and
17,400 ppm each of lindane (hexachlorocyelohexane), pentachlorophenol, and (
9,9'-dichlorofluorene (PNA) dissolved in a typical pump oil. The waste was
solidified using 19 wt percent Portland cement (type I), 17.5 percent fly ash
(class F), 37.7 percent water, 3 percent attagel 150, 1.7 percent mlcrocel E,
0.1 percent Sparr 80 emulsifier, and 0.02 percent Trlbutyl PO, antifoam so
that a total waste loading was 20 percent by weight. Leaching of the solid
after 90-day cure in 100-percent humidity using the MCC-1 static leaching
procedure (Materials Characterization Center 1981) gave saturation values
after 28 daye of leaching of <1.5 ppb PCBs, <0.76 ppb lindane, less than
10 ppm pentachlorophenol, and 0.607 ppa dichlorofluorene. The authors state
that by tailoring and adjusting the mixture, satisfactory solidification of
organic wastes in these categories (PCBs, chlorinated cyclohexane, chlorinated
phenol, and PNAs) should be possible. "
c
Pozzolanie System!
Limited Information is available on the effects of organic* in pozzolanic
system*. Researcher* have suggested that unpyrolized organic* reduce the
cementing ability of fly ash by taking up reactive surface site* of the ash
and preventing cemeuting-type contact of the variou* constituent* (Davidson
ee al. 1958, Leonard and Davidson 1959, Thorn* and Watt 1965, Thornton et al.
1975). Thornton et al. (1975), however, did *how that pozzolan* could be used
to solidify clay soil with an organic content of 11.4 percent. The addition
of * 5-percent lignite fly ash to the soil increased the UCS approximately
40 percent.
The compound* were edded up to 8.5 percent of the dry weight of the
reagent*; the formulation* used are shown in Table 16. The organic* rang* in
water solubility from Very high to virtual insolubility and cover a structural
range from simple single carbon alcohol to linear organic acid*, to simple and
substituted aromatic ring, to long unsaturated hydrocarbon chain*.
Unconfined compre**iv* strength* were mad* on 40-gram minicylindars
(1 in. diameter by 2 in. height) that were prepared on a Carver hydraulic
press (500 psi (3.4 MPa) for 20 sec). During compaction, all 8.5-p*rc*nt
organic mixe* exuded coma of the organic material. Only the adipic acid at
both 1- and 8.5-percent formulation* appreciably affected th* compr**»iv*
strength afttr 2 or 3 month*. Adipic acid is a 6-carbon, dicarboxylic acid
that must neutralize the added lima (producing calcium adipate) and therefore
interfere with th* pozzolanic reaction. Al«o evident i* th* action of
methanol a* a retarder of th* pozzolan reaction. But typical of mo*c
retarder*. th* ultimata *tr*ngth of th* mixtur* eventually equal* that
attained by th* control. Such a d*lay in setting tim* i* not generally con-
sidered to be important in most waste S/S processes. Smaller delay* in
setting tim* are also evident for xylene and benzen* at 8.5 percent, th*
effect of benz*n* *till evident (but not significant) after 90 day* of cur*.
43
-------
TABLE 16.
EFFECTS OF SELECTED ORGANICS ON UCS AND LEACHING CHARACTERISTICS OF A
LIME/FLY ASH FORMULATION
Fonuilatlona*
Z dry weight
Fly ash
95.2
95.1
94.2
67.1
95.1
94.2
87.1
95.1
94.2
87.1
95.2
94.2
87.1
94.3
87.1
Live
4.8
4.8
4.8
4.4
4.8
4.8
4.4
4.8
4.8
4.4
4.8
4.8
4.4
4.8
4.4
Additive*
None
0.1 Methanol
1.0 Methanol
8.5 Methanol
0.1 Xyleoe
1.0 Xylene
8.5 Xylene
0.1 Benzene
1.0 Benzene
6.5 Benzene
None
1.0 Adipic acid
8.5 Adipic acid
1.0 Oil * grease
8.5 Oil I grease
Unconflned coapresslvel
strength! ,
(days curing 6 'i* *"*'
7 14 28^.
190
223
219
37
257
247
201
246
245
196
185
62
57
165
161
315
305
301
41
310
308
297
313
301
267
279
95
69
271
314
454
371
435
200
446
421
362
435
452
317
290
117
66
351
388
,62 92
579
495
542
510
557
535
497
525
540
415
426
218
70
515
466
664
570
618
648
666
655
640
637
621
553
401
187
77
485
488
pH
11.65
11.70
11.55
11.90
11.70
11.70
11.85
11.65
11.70
11.90
ND
ND
ND
ND
ND
Leaching teat resultsf
28-day cure
Ca TDS
225
215
140
480
210
205
355
150
205
410
NO
ND
ND
ND
ND
890
836
564
1046
746
726
900
606
768
1030
ND
ND
ND
ND
ND
TOC
29
40
173
1538
40
44
62
29
39
41
ND
ND
ND
ND
ND
^-^ ^ ^^«,
^»^ ^
yu-day cure
pH Ca TDS
11.40
11.15
11.35
11.40
11.40
11.45
ND
11.40
11.40
ND
12.30
12.10
10.20
12.10
12.10
81
71
63
100
71
63
ND
61
81
ND
ND
ND
ND
ND
ND
728
714
712
762
724
652
ND
706
626
ND
ND
ND
ND
ND
ND
H^B^^*
TOC
8
10
87
734
10
27
ND
10
9
ND
ND
ND
ND
ND
ND
Source: Roberts 1978. Smith 1979.
Note: ND - no data; TDS - total dlaaolved solids; and TOC - total organic carbon.
All mixes prepared at 82-percent solids (including organlcs).
t All values given in pounds (force) per square Inch and are an average of three determinations.
convert values to Mgapaacala, Multiply by 0.00089457.)
f All values (except pH) are in Milligrams per litre of leachate (from ASTM 48-hour leaching test).
(To
-------
Selected leaching results upon the minicylinders after 28 and 90 days are
also shown in Table 16. The pH, total dissolved solids (IDS), calcium (Ca)
and total organic carbon (IOC) content of the leachates increased in the
8.5-percent mixes of methanol, xylene, and benzene in the 28-day cure test.
Although not shown in Table 16, the loss of methanol is quite high from the
8.5 percent mixes and remains so after 90 days of curing; these losses
represent about 20 percent of the added methanol in 48 hours of leaching from
the samples cured 28 days and about 10 percent after 90 days of curing. The
general tendency toward reduced loss for all constituents is evident for all
parameters analyzed.
No significant effect of the addition of methanol, xylene, or benzene is
seen for either arsenic or chromium (data not shown). In all cases* arsenic
averaged around 0.01 mg/l in the leachates from both the 28- and 90-day cured
samples. Chromium averaged around 0.06 mg/l in the 28-day cure samples and
0.10 mg/i after 90 days cure.
The X-ray diffraction analysis of all of the formulations gave little
information except that no new crystalline compounds were formed in detectable
amounts. Scanning electron micrographs indicated that the presence of the
organic compounds altered the morphological structure of the lime/fly ash
reaction products,'even at the 0.1-percent level and even though compressive
strength and leaching character were not affected appreciably.
Smith (1979) concluded that a good correlation exists between his find-
ings for the effects of organic compounds on lime/fly ash pozzolanic systems
and reported effects on the hydration of Portland cement; therefore, the large
amount of information concerning additives and Inferences on Portland cement
should also be applicable to systems using pozzolanic reactions. Some organic
wastes appear to be compatible with these systems, such as rolling mill
sludges, electroplating wastes, or oily sludges from petroleum refineries.
However, organic wastes containing hydroxyl or carboxylic acid functional
groups, such as biological wastes, paint sludges, and some solvents, can be
expected to delay or completely inhibit the pozzolanic reactions responsible
for solidification.
Summary of Organic Effects
In a recent study, Spooner et al. (1984) reviewed existing information
about the capability of S/S binders with different classes of organic chemi-
cals. Material for this study was obtained through a litsrsturs review and
through contacts with representatives from universities, industry, trade asso-
ciations and government agencies. From the gathered information, matrices
vere developed which summarized and defined the compatibilities of various S/S
binder systems with particular chemical groups.
The binder categories chosen for Spooner1a study include: Portland
cement types I, II, and V; clay (bentonite); clay-cement; and silicates. The
chemical categories chosen wers divided into 16 groups that reprtssntad the
typss of organic compounds most likely to be found in hazardous wastes and in
leachate from a hazardous waste disposal sits. Table 17 (Spooner et al. 1984)
was' developed from the data gathered. It presents a matrix of information
45
------- |