Jan 1, 1982 - procedures for settlement prediction of decomposing paper mill ... variability of sludge composition from site to site and the complex interaction.
16792
JANUARY 1982-
Gn
DENSIFICATION OF PAPER MILL SLUDGE MATERIALS a By Orlando B. Andersland,' M. ASCE, Abdul-Amir W. N. AI-Khafaji,l A. M. ASCE, and Richard K. Lowe J ABSTRACT: Dewatered paper industry primary sludges have a consistency similar to very soft remolded clays . Densification involves removal of additional water, usually by placement of a surcharge load after deposition in a landfill. Solids compositlOn includes a noncombustible fraction , primarily kaolin clays, and the organic constituents which include wood fibers and other components . Use of kaolinite-pulp fiber mixtures to simulate these sludges permitted control over organic content, fiber type and size, and mineral type. Decomposition was accelerated by the addition of nutrients and seed microorganisms. Use of the ignition test to determine current organic fractions permitted changes in the degree of decomposition to be monitored concurrent to changes in the model sludge compressibility. The effects of organic content, pressure, and decomposition on sludge densification was shown by compression tests. Experimental data permitted development of a method for computation of void ratio as a function of pressure and organic content. Excellent agreement was observed with limited field data. Using the degree of decomposition, procedures for settlement prediction of decomposing paper mill sludges were formulated . Examples are included. INTRODUCTION
Solids accumulated in the treatment of wastewaters of pulp and paper origin are those lost from the paper making process and subsequently separated during primary clarification. These solids are composed of fiber, filler, and coating clays. For those sludges with ash contents exceeding 50%, incineration is not a realistic alternative because of the necessity for further land disposal of the inorganic fraction . Costs associated with separate incineration of sludges in quantities commonly found at mills have largely limited the practice to those integrated operations with wood fuel boilers (9). Because land available for solid waste disposal is limited, it is desirable to optimize landfill disposal by the application of sound engineering principles to placement and densification of the waste materials. Paper industry primary sludges are characterized by their high water content, ' Presented at the May 11-16, ASCE International Convention and Exposition, held at New York, N.Y. (Preprint 81-022). I Prof. of Civ. Engrg. , Michigan State Univ. , E. Lansing, Mich. 2 Asst. Prof. of Civ . Engrg. , Univ. of Detroit, Detroit, Mich. 3 Asst. Project Engr., Soil Testing Services of Wisconsin, Inc., Green Bay, Wisc. Note.-Discussion open until June I, 1982. To extend the closing date one month, a written request must be filed with the Manager of Technical and Professional Publications, ASCE. Manuscript was submitted for review for possible publication on March 19, 1981. This paper is part of the Journal of the Geotechnical Engineering Division, Proceedings of the American Society of Civil Engineers, ©ASCE, Vol. 108, No. GT1, January, 1982. ISSN 0093-6405/82/0001-0001/S01.00.
2
JANUARY 1982
GT1
90-98 % of the sludge mass. Dewatering produces a sludge cake with a consisten.cy similar to very soft clays. Densification involves removal of additional water, usually by placement of a surcharge load after deposition of the sludge materials in a landfJll. Solids composition includes the noncombustible fraction, primarily kaolin clays with small amounts of titanium oxide and silicates, and the organic constituents which include wood fibers with small quantities of other components. To simulate paper mill sludges and to include control over organic content, fiber type and size, and mineral type , kaolin clay and pulp fibers were used to prepare model sludges. Decomposition was accelerated by addition of seed microorganisms and nutrients to the fiber-clay slurry, in proportions equivalent to those found in an average bacterial cell. Between physical tests, samples were stored at a temperature close to 35° C, conducive to decomposition . Compression tests on the model sludge materials provided data on the effect of organic content, pressure , and decomposition on compressibility. Use of the ignition test to determine current organic fractions permitted changes in the degree of decomposition to be monitored concurrent to changes in the degree of compressibility. The experimental data permitted development of relationships between the organic fraction, pressure, change in organic solids content with decomposition, and the equilibrium void ratio. Using initial and final void ratios with soil mechanics theory permitted prediction of volume change and settlement. Excellent agreement was observed between decomposed laboratory sample heights and predicted heights . PAPER MILL SLUDGE MATERIALS
Solid wastes resulting from the pulping and paper making processes include a mineral fraction and organic constituents. The sludge cake composition determines physical properties and compressibility behavior which are of interest to the engineer in planning successful and efficient high-ash paper mill sludge landfills. Composition, physical properties, and compressibility of paper mill sludge materials are briefly reviewed . Composition.-Paper industry sludge composition depends on the type of mill. Sludges with ash contents exceeding 50% are commonly associated with the manufacture of board, de-inked pulp paper, and integrated and nonintegrated fine papers (9). In addition, sludge ash has a considerable daily variation dependent upon the paper grade manufactured and performance of fiber recovery systems. Incineration is not a realistic alternative for these sludges because of their relatively low organic content, the low moisture requirements necessary for self-supporting combustion, and the necessity for land disposal of the mineral fraction . The variability of sludge composition from site to site and the complex interaction of those sludge constituents known to influence the mechanical properties require their measurement for use in the design of a landfill. The high water content of paper industry primary sludges is important relative to solids handling and disposal. Gehm (6) classified water in the sludge case as free, interstitial, or water of imbibition. Water of imbibition is chemically bound within the lattice structure of colloidal solids, and represents less than 2% of the water associated with cellulosic residues . The interstitial water held in the pores of the sludge cake by surface phenomena represents about 90% of the water associated with a thickened sludge. The remaining free water
GT1
PAPER MILL SLUDGE
3
constitutes only a small fraction of the sludge cake. The noncombustible fraction consists primarily of filler materials, such as kaolinite and titanium dioxide, with possible quantities of silicates and carbonates (9) . Their particle size is predominantly smaller than 2 IL. The organic fraction may consists of fiber and such colloidal components as highly hydrated wood dust, fiber debris, ray cells, starches, dextrines, resins, and protein. The fibrillar structure of cellulosic components and the size of interstices impart a high capacity for retention of water by capillary forces. Length and diameter of the cellulosic constituents influence both the dewatering process and mechanical properties. Physical Properties.-Water (or solids) content, unit weight, specific gravity, and mineral and organic fractions are properties useful in describing the engineering characteristics of paper mill sludge. Available American Society for Testing and Materials (ASTM) standard procedures include D 2216 for laboratory determination of soil moisture content and D 854 for specific gravity of the sludge solids. For an experimental high-ash paper mill sludge (3) water contents ranged from 308 % during placement of sludge cake to as low as 142% after consolidation. The equilibrium water content is highly dependent on the amount of organic material present. Specific gravity of sludge materials in an experimental landfill (13) ranged from 1.87-2.24 . Higher values correspond to those sludges with higher mineral fractions . The test for specific gravity requires special care so as to minimize the effect of entrapped air and possible formation of gas bubbles during the test as a result of biological decomposition. The bulk unit weight of fresh sludge, defmed as the weight of the sludge and water per unit volume, can be determined using a container with known volume. Careful placement of sludge into the container is required to minimize the effects of any air pockets. At some sites, a truck transporting the sludge can be weighed both when full and when empty from which the bulk unit weight may be computed using the volume and weight of sludge. Unit weights reported for an experimental high ash paper mill sludge landfill (3) ranged from 10.8 kN/m 3 for fresh sludge cake to about 12.2 kN/m 3 after consolidation at a depth of about 7 m. Mineral and organic fractions help characterize the sludge cake relative to equilibrium water contents and compressibility. Recent information on measurement of the mineral and organic fractions is given later in this paper under the heading ignition test. Compressibillty.-Paper mill sludge densifies or consolidates under load in a manner similar to organic soils. Settlement-time curves for an experimental paper mill sludge landfill are shown in Fig. 1. Sand drainage blankets are shown at the top and bottom of two sludge layers, each initially 3 m thick. Decrease in thickness of these layers, monitored using settlement plates, was rapid immediately after sludge placement and then decreased to rates corresponding to secondary compression. Vallee (13) evaluated the field coefficient of consolidation for an incremental surcharge load by plotting settlement versus the square root of time. An additional soil surcharge would increase the settlement until new equilibrium void-ratios were established in the sludge. If conditions are suitable for decomposition, additional settlement occurs as organic material is converted into a humus-like material with a decrease in volume of organic solids. Gases, water, volatile acids, and new microbial cells are by-products of decomposition.
4
GT1
JANUARY 1982
Conventional methods used to estimate settlement include compression tests to determine a relationship between void ratio and pressure. Decrease in volume is then computed on the basis of change in void ratio for known stress conditions. A common method involves use of a compression index dermed in terms of the change in void ratio per increment of logarithm of pressure. For inorganic soils, this compression index remains reasonably constant over the range of working stresses. For soils with high organic contents this is not the case (4). Organic material with its high water holding capacity significantly alters the material behavior. Methods for prediction of volume change caused by loss of organic solids as a result of decomposition have not been available. New procedures have required accurate determination of organic contents so that
~
Ow Q~
~ »
w
SOI L
SURCHAR GE
El ~~~ ~L:~:~~ E~L:~~H
~
IZl 2
wo
-"" w«
SLUDG E
I NSTRUMENT
120
GROU P
150
4
180
2 10
TIME, da ys
FIG. 1.-Settlement-Time Curves for Experimental High Ash Paper Mill Sludge Landfill (13)
the degree of decomposition could be more precisely known for computational purposes. These topics are discussed in later sections of this paper. EXPERIMENTAL WORK
The experimental work involved preparation of model sludges followed by consolidation tests . A brief description of the materials studied, sample preparation, ignition test, and compression tests are included. Materials Studied.-Pulp fibers included a range of sizes with a weighted average length of 1.6 mm, based on the test procedure of the Technical Association of the Pulp and Paper Industry (12) and typical diameters of about 20 IJ.m observed using an electron microscope . Surface area measurements (7), using the water vapor absorption method (11), gave values close to 133 m 2 /g of dry fiber . A value of 1.54 was used for specific gravity of the pulp fiber. The kaolin clay included particles with 100% passing 40 IJ.m, 93 % passing
GT1
PAPER MILL SLUDGE
5
10 IJ.m, and 42% passing I IJ.m. The grain size distribution was relatively well graded. Consistency limits included a liquid limit of 47 .8, plastic limit of 27.5 , and plasticity index of 20.3 . The specific gravity of kaolin solids was equal to 2.70. Sample Preparation.-Pulp fiber board, after freeze drying, was separated into a fluffy mass. Selected proportions of dry fiber and dry kaolinite were mixed. Next distilled water was added in amounts needed to form a slurry. Nutrients and seed microorganisms, in predetermined proportions, were added directly to the slurry for those experiments involving decomposition. Nitrogen served as the base by which other nutrient quantities were selected. Approximate ratios giving optimal decomposition rates included carbon / nitrogen, C / N = 30:1; phosphorus/nitrogen, PIN = 1:5; and (Mg, Fe, K, Ca, and Na)/N = 1:16. Pulp fiber supplied carbon, NH 4CI supplied nitrogen , and other source compounds included K 2HP0 4 for phosphorous and potassium, MgS04 for magnesium, CaCI 2 for calcium, and FeCI) for iron . About I % (dry weight basis) of municipal sludge provided seed microorganisms. These ratios appeared to satisfy the microorganism nutritional requirements (10) . Ignition Test.-Measurement of the sludge organic fraction requires that it be separated from the mineral solids. For engineering purposes, ignition at high temperatures is the most common method. Destruction of the organic component by ignition requires that other sludge constituents not be altered so that the weight loss, compared to that at 105 ° C, can be taken as a measure of organic content. AI-Khafaji (I) prepared weight loss reduction curves which showed complete combustion for pulp fiber at temperatures below 400° C and that surface hydration moisture loss for kaolinite was dependent on ignition temperature . The organic fraction , X f , is given by Xf
=I-
W2 Xm = I - C WI
. .. . . . . . . . .. . . . . . . . . . (I )
in which Xm = the mineral fraction ; WI = the initial oven dry weight at 105° C; and W2 = the fmal sample weight after ignition. The constant, C, depends on the ignition temperature. AI-Khafaji (I) determined that C = 1.014 at 400° C (12 hr ignition) or C = 1.168 at 900° C (1.25 hr ignition) gave accurate values for the model sludge organic fraction. Test apparatus, sample size, and procedures for determination of the mineral fraction were similar to those described in the standard test method for ash in paper (ASTM Designation D586-63). Compression Tests.-One-dimensional compression tests (7) for stresses ranging from 0. 12 MN/m 2-35 MN/m2, were conducted on model sludges containing no nutrients using special compression test cyclinders with sample cross-sectional areas of 100 and 645 mm 2. Cylinder heights were sufficient to permit use of model sludges with an initial slurry consistency. Water drainage through top and bottom porous stones provided a check on decrease in sample height. A force transducer mounted below the bottom porous stone permitted load measurements to an accuracy of ±5 N. Primary consolidation, as defined by the square root fitting method (8), was complete in less than 30 min for no decomposition. Each load increment was mainta~ned for a 24-hr period. Slurry samples containing nutrients and seed microorganisms were consolidated in 3 L beakers with all drainage through the top porous stone. Consolidation stresses of 0.47 , 1.14, 2.28, and 3.42 kN/m 2 were used for the initial condition
6
GT1
JANUARY 1982
(no decomposition) and later stages of partial decomposition. The second and third load increments were started as soon as the previous load increment reached 100% primary consolidation as dermed by the square root fitting method (8). Drained fluids were returned to remolded samples after each compression test series so that anaerobic decomposition could continue at an accelerated rate during storage at a temperature close to 35 0 C. EOUIUBAIUM VOID RATIO
Density of paper mill sludge materials is dependent on particle packing and eqUilibrium void ratios for existing loads, sludge composition, temperature, and PRESSURE kN/m2
:> 12
10
.,
200 600 1500 8000
~}DATA •
INTERCEPT
SLOPE
1.20 1. 14 1.07 0 .89 0 .68
13 .06 2 .60 1.32 0.55 0.07
FROM
DATA FROM AL-KHAFAJI
8
0
!i 0:: 0
6
0
> 4
8000 kN/m2
00
0 .2
0 .4 0.6 ORGANIC FRACTION,
0 .8 Xf
FIG. 2.-Relationships Between Organic Fraction. X f • Void Ratio. e. and Consolidation Pressure. p. for Model Sludge Samples
chemical environment. Model sludge behavior during applicaton of an external load, void ratio prediction in terms of the organic fraction and pressure, comparisons with field data, and settlement prediction are described in the following sections. Model Sludge Behavior.-Application of an external stress to paper mill sludge materials causes a gradual decrease of water content until sufficient shearing resistance is developed through interparticle contacts to resist interparticle shear forces. The presence of organic material, with greater water holding capacity, . increases the equilibrium water contents and corresponding void ratios at lower
7
PAPER MILL SLUDGE
GTl
stress levels. The behavior of model paper mill sludges, for one-dimensional compression and 24-hr loading periods, is summarized in Fig. 2 where organic fraction has been plotted against void ratio for selected pressure levels. Temperature was held constant during compression tests and no decomposition was permitted. An organic fraction of zero corresponds to all kaolin clay and a value of one to all fibers . At each pressure level a straight line gives a good representation of the experimental data. At high pressures, greater than 10 MPa, the void ratio was essentially unaffected by organic content. An equation may be written for each line giving the void ratio, e, in terms of the pressure, p, and the organic fraction, Xf , thus e(p, Xf
)
= C(p) + M(p) Xf
. ... . .. . . . . . ... .. ... . . .. . .
(2)
in which C(p) = the intercept and M(p) = the slope. The intercepts and slopes appear to be unique to their respective consolidation pressure and are listed on Fig. 2.
12
"
10
. Ip. X,), Clp )
+ Ml p) X,
X f '" I - I. 168 Xm
where
X f " organic frac ti on by we1o ht,
Xm= mine ral fr act io n by weio nt base d on ig ni ti on at 900"e f or 1. 2 5 hr .
.. 6r"---~------
