vxEPA
United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/2-78-111
May 1978
A Study of the
Feasibility of
Utilizing Solid
Wastes for Building
Materials
Phase III and IV
Summary Reports
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1 Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-78-111
May 1978
A STUDY OF THE FEASIBILITY OF UTILIZING SOLID WASTES
FOR BUILDING MATERIALS
Phase III and IV Summary Reports
by
Material Systems Corporation
Escondido, California 92025
Contract No. 68-03-2056
Project Officer
Robert E. Landreth
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation
for use.
11
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health and
welfare of the American people. Noxious air, foul water, and spoiled land are
tragic testimony to the deterioration of our natural environment. The com-
plexity of that environment and the interplay between its components require
a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solution,
and it involves defining the problem, measuring its impact, and searching for
solutions. The Municipal Environmental Research Laboratory develops new and
improved technology and systems for preventing, treating, and managing
wastewater and solid and hazardous waste pollutant discharges from municipal
and community sources, for preserving and treating public drinking water
supplies, and for minimizing the adverse economic, social, health, and aesthe-
tic effects of pollution. This publication is one of the products of that re-
search, a most vital communications link between the researcher and the user
community.
This project has resulted in savings for the environment and man. En-
vironmental stresses are reduced when waste streams can be productively sub-
stituted for natural resources. These building products can benefit man in
the areas of both safety and economy. Unique solutions such as these are to
be encouraged.
Francis T. Mayo
Director
Municipal Environmental
Research Laboratory
Xll
-------
ABSTRACT
This report summarizes work conducted during Phases III and IV of a study
by Material Systems Corporation to develop building materials containing
organic and inorganic wastes and waste-derived products.
The first phase, April 1974 to July 1974, included an extensive litera-
ture search to identify, review, and evaluate wastes with potential as
matrices, reinforcements, or fillers in building composites. Following limited
laboratory studies, two matrices were selected—an organic resin and an inor-
ganic matrix developed by Material Systems Corporation. Seven reinforcement
candidates and five fillers were chosen for detailed evaluation.
The second phase, July 1974 to September 1975, was an evaluation of the
structural and aesthetic properties of the various waste composites. Satis-
factory products were made using both inorganic and organic systems; structural
and fire-resistant properties of several formulations were found to be supe-
rior to those of existing commercial products. Economic analyses were per-
formed on the structurally promising products, and several appear to be com-
mercially viable.
In Phase III, September 1975 to November 1976, attempts were made to pro-
duce full-scale products and qualifty them for structural applications. Parti-
cle board panels, 4 by 4 feet, were made of peanut hulls and wood waste on
production-type equipment. Particle boards of peanut hulls have mechanical
properties that are slightly less desirable than those of commercially avail-
able boards, and the economics are marginal. However, particle board panels
of wood waste can be competitive with commercial products. Two-hour, fire-
rated structural walls made from inorganic rice hull foam could also be vi-
able, as could floors, roofs, ceilings, and the 90-minute, fire-rated door
with a rice-hull foam core and a wood-waste frame. Data were developed for
submittal to the International Conference of Building Officials and the U. S.
Department of Housing for certification.
The components were incorporated into an 8- by 12-foot demonstration mod-
ular unit and erected at the research center in Cincinnati, Ohio.
In Phase IV of the study, structural tests were performed on wall panels
fabricated from rice hulls and an inorganic binder. These tests completed
generation of the data required for building code approval.
Wall panels made of rice hull composite were found to satisfy all
structural requirements necessary for use in one-and two-story residential
construction, except that resistance to transverse loading (bending) was
insufficient.
IV
-------
Long-term weathering exposure tests of the rice hull composite indicated
high sensitivity to moisture degradation. The development of protective coat-
ing materials that would prevent moisture penetration and increase the bending
resistance of the wall panels is recommended for future work.
Rice hull wall was demonstrated to withstand long fire exposure without
structural failure. A wall panel assembly was fully qualified for 1-hour
structural fire rating.
The most successful accomplishment of this study was the development of
new core and framing materials for fire doors. In an exemplary effort of
cooperation between a research organization and a manufacturer of commercial
products, new materials were developed for specific application. They were
qualified in full-scale tests to meet building code standards and were
adapted for economical production. As a result, the manufacturer is planning
to introduce into his product line a high-performance, 90-minute fire rated
door using materials based on waste products and developed during this study.
This report was submitted in fulfillment of Contract No. 68-03-2056 by
Material Systems Corporation under the sponsorship of the U. S. Environmental
Protection Agency. This report covers the period December 1976 to February
1977, and work was completed as of February 1977.
-------
CONTENTS
Foreword ill
Abstract iv
Figures viii
Tables x
Metric Conversion Factors xii
Phase III
1. Introduction 1
2. Waste Products Using an Organic Matrix 3
3. Waste Products Using an Inorganic Matrix 77
4. Roof and Ceiling System 117
5. Floor System 130
6. Fire Door 134
7. Demonstration Unit Constructed from Waste Material 149
Appendices
A. Information sources on the economics of particle board. 152
B. Data on water effects on rice-hull reinforced
inorganic matrix 153
Phase IV
1. Introduction 163
2. Structural Tests of Wall Panels 165
3. Fire Test of Wall 174
4. Fire Door Development 175
VII
-------
PHASE III FIGURES
Number Page
1-1 Forming mat 36
1-2 Placing board on mat 36
1-3 Covering with Mylar 37
1-4 Placement of aluminum caul plate 37
1-5 Insertion into press 38
1-6 Molding 38
1-7 Test specimen cut-up diagram 39
1-8 Internal panel temperature 44
1-9 Cracking phenomenon 45
1-10 Steam generation 45
1-11 Steam generation 47
1-12 Panel removal 47
1-13 Panel removal 48
1-14 Panel cooling 48
1-15 Section cut for test. 49
1-16 Panel explosion 49
1-17 Debris from explosion 50
1-18 Segment of exploded panel 51
2-1 Effects of water on strength retention 80
2-2 Percent strength comparison of rice hull/inorganic and wood 81
2-3 Cross section of Phase II wall 83
2-4 Full-scale Phase II wall 84
2-5 Cross section of wall material 86
2-6 Rice hull foam wall elements 87
2-7 Tooling for scale-up studies 87
2-8 Assembly of wall material into a panel 88
2-9 Cross section of specimen 89
2-10 Compression failure 90
2-11 Material Systems Corp. panel axial compression test set-up 91
2-12 Full-scale compression test 92
2-13 Compression load vs deflection and permanent set
for a 4 ft x 8 ft rice hull foam fire wall 94
2-14 Wall specimen 95
2-15 Material Systems Corp. panel racking shear test 96
2-16 Racking shear test on panel without reinforcement 97
2-17 Racking shear load vs deflection and permanent set 99
2-18 E-119 fire wall test 102
2-19 Small section fire test 103
2-20 Mounting walls into test frame 104
2-21 Instrumentation 104
2-22 Buckling during endurance test 105
2-23 Stiffener bond failure to unexposed surface 106
viii
-------
Number
2-24 Exposed surface upon removal from furnace 106
2-25 Hose stream test 108
2-26 After hose stream test 1C8
2-27 Panels with new rib 109
2-28 Three panels ready for shipment 109
2-29 Calorimeter Ill
2-30 Floor and ceiling specimen 118
2-31 Beam-bending test 119
2-32 Beam-bending vs deflection 120
2-33 Beam-bending test (long-term loading) 121
2-34 Long-term loading test: Uniformly loaded 21.9 lbs/ft2
beam-bending dry 122
2-35 Uniformly loaded 10.9 lbs/ft2 beam-bending running water 124
2-36 Elements for a floor panel 129
2-37 Fire door design 145
2-38 Surface after 90-min fire and hose stream 146
PHASE IV FIGURES
Number Page
1 Transverse load vs deflection curves for rice hull-inorganic
matrix wall panels 166
2 Racking shear load vs deflection curves for rice hull-inorganic
matrix wall panels nailed to top and bottom plates 167
3 Long-term exposure test fixtures 169
4 Cyclic environmental chamber 170
5 Axial and lateral deflection of rice hull-inorganic matrix
wall panels during long-term rain soak exposure under
constant axial load of 250 Ib/ft 171
6 Axial compression load vs. deflection curves for 4 by 8-ft
rice hull-inorganic matrix wall panels after various
long-term exposures 172
IX
-------
PHASE III TABLES
Number ' Page
1-1 Composition by % Weight of Peanut Hulls-Phosphoric Acid Mixture. . 5
1-2 Titration Studies of Phosphoric Acid Soak Solutions With Reuse . . 5
1-3 Acid Used in Large-Scale Mixer 6
1-4A Small Peanut Hull Panels Made With Phosphoric Acid 6
1-4B Medium-Sized Peanut Hull Panels Made With Phosphoric Acid .... 8
1-5 Processing Studies on Ponderosa Pine 10
1-6 Processing Studies on Ponderosa Pine Using Mixed Acid 11
1-7 Processing Variables, Treatment of Hard Pine Wood Using
Phosphoric Acid • ^
1-8 Processing Variables, Treatment of Douglas Fir 13
1-9 Processing Variables, Treatment of Douglas Fir 14
1-10 Processing Variables, Treatment of Douglas Fir 15
1-11 Commercial Douglas Fir (-17 Mesh) With P-Toluene Sulfonic Acid .. . 17
1-12 Commercial Douglas Fir (+17 Mesh) With P-Toluene Sulfonic Acid . . 18
1-13 Commercial Douglas Fir With Ultra Tx Acid 19
1-14 Longview Douglas Fir (20 Mesh)/P-Toluene Sulfonic Acid 20
1-15 Longview Douglas Fir (+20 Mesh)/P-Toluene Sulfonic Acid Spray. . . 22
Method 22
1-16 Effect of Molding Time on Properties 24
1-17 Effect of Molding Time on Properties 25
1-18 Effect of High Press Temperature and Shorter Press Times
on Properties of Peanut Hull Waste Panels 26
1-19 Effect of Post Curing Conditions on Peanut Shell Properties. ... 28
1-20 Effect of Postcure on Peanut Hull Panels 30
1-21 Effect of Postcure on Peanut Hull Panels 30
1-22 Effect of Postcure on Peanut Hull Panels 31
1-23 Effect of Postcure on Peanut Hull Panels 31
1-24 Use of Preheated Peanut Hulls Molded at Various Press Cycles ... 33
1-25 Use of Preheated Peanut Hulls Molded at Various Press Cycles ... 33
1-26 Use of Preheated Peanut Hulls Molded at Various Press Cycles ... 34
1-27 Use of Preheated Peanut Hulls Molded at Various Press Cycles ... 34
1-28A Pressing Conditions and Averaging Physical Properties of the
50 by 50 Inch Panels 41
1-28B Pressing Conditions and Averaging Physical Properties of the
50 by 50 Inch Panels 42
1-29 Cut-Up Diagram and Physical Test Results of Board Made With
Urea Formaldehyde Resin (Peanut Hull Furnish) 52
1-30 Board-Making Conditions for Peanut Hull Furnish 54
1-31 Board-Making Conditions for Peanut Hull Furnish With Urea Resin. . 56
1-32 Cut-Up Diagram and Physical Test Results for Board 27
(Peanut Hull Furnish) 58
1-33 Board-Making Conditions for Douglas Fir Furnish 60
-------
aqe
Number
1-34 Cut-Up Diagram and Physical Test Results for Board 66
(Douglas Fir Furnish) 64
1-35 Chip Board Product Using Douglas Fir Fines With P-Toluene
Sulfonic Acid 75
2-1 Rice Hull Foam Formulation 77
2-2 Cured Weight and Cost Distribution 78
2-3 Mechanical Properties of Rice Hull Foam 82
2-4 Cost of Waste Product Two-Hour, Fire-Rated Partition Wall .... 113
2-5 Magnesium Oxy-Chloride Coating Cost/Square Foot 114
2-6 Cost of a Conventional Two-Hour,'Fire-Rated Partition 114
2-7 Cost of a Conventional Interior Wall 115
2-8 Cost of a Conventional Exterior Wall 115
2-9 Cost of Waste Product Roof-Ceiling Panel With Hardware Cloth
and Metal Lathe Reinforcement 126
2-10 Cost of Conventional Roof-Ceiling System 127
2-11 Cost of Waste Product Ceiling Panel 127
2-12 Cost of Conventional Ceiling System 128
2-13 Cost of Waste Product Floor-Ceiling Panel With Metal Lathe
Reinforcement 132
2-14 Cost of Magnesium Oxy-Chloride Floor Surface 133
2-15 Cost of Conventional Floor-Ceiling System 133
2-16 Cost of Conventional Wood Floor System 133
2-17 RH-54 Formulation 135
2-18 Fire-Door Insulation Material Formulations Based on Rice Hulls. . 136
2-19 Attempts to Case Full Size Fire Door Core (34.5" x 77" x 1.5") . 139
2-20 Formulation for Fire Door Core 139
2-21 Comparison of Matrix Materials for Door Framing Using
Phenolic Microballoons 141
2-22 Comparison of Matrix Materials for Door Framing Using Silvacon. . 141
2-23 Use of Various Waste Fillers for Door Framing Using
Magnesium Oxy-Chloride 142
2-24 Properties of Various Woods Used in Door Frames 145
2-25 Formulation for Framing Material 145
2-26 Fire Door Core Economics 148
2-27 Framing Material Economics 148
B-l Uncoated: Rice Hull Reinforced Inorganic Matrix—Weight Data. . . 153
B-2 Uncoated: Rice Hull Reinforced Inorganic Matrix 154
B-3 Coated Zynolite Paint: Rice Hull Reinforced Inorganic Matrix
(Coated)—Weight Data 155
B-4 Coated Zynolite Paint: Rice Hull Reinforced Inorganic Matrix
(Coated) 156
B-5 Coated Lytron 621: Rice Hull Reinforced Inorganic Matrix
(Coated)—Weight Data 157
B-6 Coated Lytron 621: Rice Hull Reinforced Inorganic Matrix (Coated) 158
B-7 Coated W-102: Rice Hull Reinforced Inorganic Matrix
(Coated)—Weight Data 159
B-8 Coated W-102: Rice Hull Reinforced Inorganic Matrix (Coated). . . 160
B-9 Kiln-Dried Douglas Fir Lumber—Weight Data 161
B-10 Kiln-Dried Douglas Fir Lumber 162
XI
-------
METRIC CONVERSION FACTORS
All measurements in EPA documents are to be expressed in metric units.
In this report, however, this practice adversely affects clarity. Conversion
factors are therefore given below
METRIC CONVERSION TABLE
English system
Metric system
Inch
Foot
Yard
Square inch
Square foot
Square yard
Cubic inch
Cubic foot „
(1,728 in )
Cubic yard
(27 ftJ)
Pound
avdp.
troy
(16 dr)
(12 oz)
Ton
long (2,240 Ib)
short (2,000 Ib)
2.54 cm
30.48 cm
0.914 m
6.452 cm2
0.092 m2
0.836 m2
16.387 cm3
0.028 m3
0.7646 m3
-17°C
453.592 g
373.24 g
1.016 t
0.907 t
xix
-------
PHASE III SUMMARY REPORT
SECTION 1
INTRODUCTION
Material Systems Corporation and the Solid and Hazardous Waste Research
Division of the Municipal Environmental Research Laboratory at Cincinnati,
Ohio have initiated a joint effort development study for the application
of waste materials for residential and commercial construction. This is
a summary report of the Phase III activity. Phase I activity was reported
in MSC Technical Report 74-12. This report presented a comprehensive
literature search.
The data were reviewed and evaluated for potential matrix, reinforcement,
and filler candidates. The more promising candidates were evaluated with
limited laboratory studies. Prom these studies, two types of matrices,
furfural derivatives and inorganic, were selected for further study.
Seven reinforcement candidated and five fillers were selected for evalua-
tion with the two matrices.
The Phase II activity demonstrated the potential technical and economic
feasibility of utilizing waste materials for building construction. As
the Phase II study progressed, it became apparent that there existed a
multiplicity of product possibilities from waste materials. Therefore,
it was jointly decided to concentrate Phase II activity on the development
of the product and postpone specific qualifications until Phase III.
The development included testing and evaluation to sufficient level to
assure probable qualification for specific products. The products selected
for qualification studies in Phase III were in two main categories: those
materials from organic matrices and those from inorganic matrices.
The Phase III study selected wood waste and peanut shells for possible
qualification as organic particle board. Panels as large as four feet by
four feet were fabricated and tested. However, not all the processing
problems were resolved which would require further studies. The wood
waste particle board has some economic advantage over commercial particle
boards. The economic advantage of the peanut shell board is marginal.
With the resolution of some of the processing problems, the boards could
be qualifiable. They do have one limitation and that is density. Boards
from this process are limited to a minimum weight of 55 pounds per cubic
foot. The details of this study are presented in the section "Waste
Products Using an Organic Matrix."
The Phase III study also selected a rice hull foam from an inorganic
matrix for the qualification of walls, floors, roof-ceiling, and floor-
ceiling systems. A satisfactory load carrying two hour fire-rated wall
-------
was developed. An acceptable floor, floor-ceiling, and roof-ceiling
system was also developed. All systems have an economic advantage over
conventional structures and lend themselves to mass production techniques.
Further studies are recommended on the problem of long term loading.
The rice hull foam above was modified to be used as a core for fire doors.
A screwable framing material from wood waste-reinforced inorganic matrix
was also developed. The resulting core and framing materials both have
economic advantages and offer a more reliable 90 minute fire door than do
those commercially available.
Details of the organic studies are presented in the section "Waste Products
Using an Inorganic Matrix."
The final activity of this study was the construction of an 8-ft. wide,
12-ft. long demonstration unit that incorporates all of the above products.
Details are presented in the section "Demonstration Unit Constructed From
Waste Products."
-------
SECTION 2
WASTE PRODUCTS USING AN ORGANIC MATRIX
The Phase II study developed a two-step process of "in-situ" formulation
of furfural from wastes such as wood, peanut hulls, rice hulls, and cotton
stock. It was discovered that materials containing cellulose and pentosans
when treated by acid, yield a binder during the process of molding. The
acid treatment is very simple. A 3% to 7% solution of hydrochloric,
phosphoric, or p-toluene sulfonic acid is used depending on the waste
involved. The first step if to soak the material in the acid solution
for 30 minutes and then dry it in an oven at 50-60 C. The second step
is to transfer the material to a cold mold and cure it under pressure for
2 to 10 minutes. This concept appeared to be a feasible application to
the production of particle board. The results of work with 4 inch by 6
inch specimens supported this. However, when a limited quantity of 30 inch
by 30 inch panels were made, scale-up problems became apparent but were
considered solvable.
The activity of Phase III was then directed to the development of process--
ing conditions which would permit the fabrication of 48 inch by 48 inch
panels that could be qualified as exterior structural particle board. The
basic conditions to be resolved were as follows:
•
1) Type and percent of acid absorbed by material
2) Moisture content
3) Time under pressure
4) Press temperature
5) Pressure technique
6) Post cure
7) Pre-heat
8) Particle size
These are discussed in the following text. Prior to initiating the study
the candidate waste materials were limited to peanut hulls and wood waste
as they provide the most resinous binder by in situ formation.
TYPE AND PERCENT OF ACID ABSORBED BY THE MATERIAL
Peanut Hull Waste--Two acids were considered for use with peanut hull
waste—p-toluene sulfonic acid and phosphoric acid. There are no apprecia-
ble differences in properties and tests indicate that a 4 to 6% acid
absorption by the hulls is the minimum acceptable for either with peanut
hulls. The cost of phosphoric acid is $0.18 per pound and the cost of
p-toluene sulfonic is $0.25 per pound. Therefore phosphoric acid was used
-------
in all peanut hull waste scale-up studies.
To acid soak large quantities of peanut hulls, it was necessary to
abandon laboratory methods which soaked the hulls in a 3% solution of 75%
acid in water for 15 minutes followed by draining. Later, a washing
machine and clothes dryer were effectively used. Bags were sewn with fine
polyester mesh fabric and hulls put into these for soaking and centrifug-
ing. Without these bags, the vents in the dryer were rapidly plugged,
drying efficiency drastically reduced, and there was also a considerable
loss of material. It was obvious that the centrifugation of the soaked
peanut hulls would remove much more of the phosphoric acid solution used
to treat the hulls, thus leaving behind less acid in the hulls. The
properties of the product were greatly affected by the amount of
phosphoric acid in the hulls. It was necessary to determine the amount
of acid in the hulls when drained by gravity in the lab method on small
quantities vs that of the centrifuged material. This was determined by
soaking hulls in acid using both methods. The results of this test are
presented in Table 1-1. It can be seen from these data that the ratio of
hulls to absorbed acid is much greater with the spin process than with
the laboratory soak process, therefore a higher acid concentration in
the solution was required to achieve equivalent acid absorption. The
moisture content was lowered by centrifugation and this was advantageous
because of more rapid drying with lower energy consumption.
Another aspect examined for scale-up was the re-use of drained acid
solution with added fresh acid solution. By this procedure a considerable
cost savings resulted compared to the labscale method where the drained
acid was discarded. Titration results are presented in Table 1-2.
After using three times, the acid soaking solution was made strongly
alkaline with sodium hydroxide solution and this produced an amine-like
order which caused indicator paper to show a pH of 9 for the vapors above
the solution. Therefore it appears that peanut hulls contain a basic
material which removes hydrogen ion by neutralization but permits reuse
up to three times.
The lab method soaking time was reduced from 15 minutes or longer to 5
minutes. This was determined by soaking peanut hulls in 3% acid solution
for 5 minutes and 800 minutes (16 hours). The initial ratio of
H1/H5 was °-98- After 5 minutes the ratio was 0.785 and after 800 minutes
it was 0.79, indicating no further change after 5 minutes.
After establishing the method of acid absorption, two large-scale mixes
were made to produce material for use in full scale process studies at
Washington State University. The solution used was 8 gallons of a 7%
solution of 75% phosphoric acid in water for 8.82 pounds of hulls. The
acid utilizationis presented in Table 1-3. These samples were further
dried at Washington State University removing more of the acid and
solution. Table 1-4A shows the results with varying acid content on 4
inch by 6 inch panels and 1-4B on 12 inch by 15 inch panels. It would
-------
TABLE 1-1 COMPOSITION BY % WEIGHT OF PEANUT HULLS - PHOSPHORIC ACID MIXTURE
Method
% Hulls
% Acid % Water
(100%
Concen-
trate
Hulls/Acid
Lab Process Test #1
Test #2
Test #3
Spin and Dry after
3% Sol of 75% Acid
in water
Spin and Dry after
7% sol of 75% Acid
in water
20.7
23.1
24.9
43.1
44.4
2.3
2.2
2.2
1.66
3.64
77.0
74.7
72.9
55.2
52.0
9.0
10.5
11.3
26.0
12.2
TABLE 1-2 TITRATION STUDIES OF PHOSPHORIC ACID SOAK SOLUTIONS WITH REUSE
Theoretical Molarity = 0.224
Item
Hi (pH 5.75)
HI (pH 8.6)
Ain't of 1st Hydrogen Used
Ara't of 2nd Hydrogen Used
O-Use
0.215 Molar
0.213 Molar
0
0
1-Use
0.161
0.218
25%
0
2-Use
0.115
0.220
47%
0
3-Uses
0.062
0.194
71%
0
-------
TABLE 1-3 ACID USED IN LARGE-SCALE MIXER
Wt of Peanut Hulls Wt. of 75% Phosphoric Acid % of 100% Ac
Hulls
1000 73.51 5.5%
1300 117.62 6.8%
TABLE 1-4A SMALL PEANUT HULL PANELS MADE WITH PHOSPHORIC ACID
Reference
Book
No. 69
30-14
30-15
30-16
30-17
30-18
38-1
v
38-2
3S--3
38-4
38-5
38-6
38-7
38-10
38-11
42-21
42-22
42-23
30-12
30-13
38-8
38-9
42-19
42-20
MATERIAL
Acid
(g/lOOg
of chips)
4.96
4.96
4.96
2.53
0.885
4.96
4.96
4.96
4.96
4.96
4.96
4.96
4.96
4.96
4.19
6.28
6.28
4.15
2.02
1.044
1.044
1.62
3.45
4 x
6 in.
Panels
MOLDING CONDITIONS
%Moist
Content
*
9.0
9.0
9.0
6.6
8.6
9.0
9.0
5.5
9.0
1.3
9.0
9.0
2.8
2.8
7.1
6.1
6.1
11.3
8.9
6.9
6.9
11.1
11.8
Mold
Temp
(°F)
377
385
418
418
415
437
445
430
350
349
377
405
432
432
350
345
370
427
432
416
435
395
379
Pressure
(p'si)
750
750
750
750
1000
1000
1000
1000
500
500
750
750
1000
1000
750
750
750
750
1000
1000
1000
1000
1000
Pressure
Relief
Cycle
(min)
None
None
3,5,7,9
3,5,7,9
None
1,3,5,7
9,11,13
2,4
6,8,10,
12,14
None
None
None
2,8
4,10
r;
None
None
None
None
None
None
14
None
6,8
Time
Under
Pressure
(min)
10
10
10
10
10
15
15
15
26
25
25
20
10
15
10
10
10
4
10
20
20
10
10
10 grams, 1 hour @ 300 F in forced-air oven.
(continued)
-------
TABLE 1-4A (Continued) SMALL PEANUT HULL PANELS MADE WITH PHOSPHORIC ACID
PROPERTIES OF MOLDED PANEL
Reference Blown Relative Density Hardness Flex-Mor(psi)
Book Area Shrinkage (Ibs/ft ) Shore D Initial 24-hr
(avg. of 3) water
soak
(avg of
2)
30-14
30-15
30-16
30-17
30-18
38-1
38-2
38-3
38-4
38-5
38-6
38-7
38-10
38-11
42-21
42-22
42-23
30-12
30-13
38-8
38-9
42-19
42-20
0
0
0
0
0
0
80
40
0
0
0
100
Expl.
Expl.
0
0
0
Expl.
0
0
0
0
100
High
V. High '
V. High
V. High
Medium
V. High
V. High
V. High
High
High
High
V. High
-
-
High
High
High
V. High
V. High
High
V. High
V. High
V. High
60
63
70
66
57
74
77
73
59
55
65
70
71
-
61
51
55
57
68
60
65
58
72
67
73
79
74
66
77
83
78
72
45
70
82
-
-
69
58
61
61
70
68
76
72
80
971
1002
1934
1810
857
1846
2677
2469
1079
664
2082
2358
-
-
1422
460
472
-
1728
1135
1170
962
1749
239
397
1445
1280
53
895
688
636
177
51
974
1555
-
-
699
22
88
-
412
88
551
211
1096
-------
TABLE 1-4B MEDIUM-SIZED PEANUT HULL PANELS MADE WITH PHOSPHORIC ACID
(12" x 15"]
MATERIAL
Reference
Notebook
#69
46-1
46-2
46-3
46-4
46-5
46-6
Acid Moisture
(g/lOOg)
of chips
3.98
3.98
3.98
3.98
3.98
3.98
WSU
Method %
5.5
5.5
5.5
5.5
5.5
5.5
Flatten
Temp
(°F)
375
375
375
340
390
390
PROPERTIES OF
Reference
Notebook
#69
46-1
46-2
46-3
46-4
46-5
46-6
Charge
Size
grams
1900
1900
1900
1900
1800
1800
Thickness
-
0.542
-
0.525
0.525
0.545
Density
lb/ft3
65
65
64
64
1
MOLDING
Pressure
psi
1000
1000
1000
1000
1000
1000
PANEL
Internal
BOND
IB psi
164
195
147
183
CONDITIONS
Pressure
Relief
Cycle
(min)
None
5
5,10
Blew
5
5
5,10
Flexure
MOR MOE
psi xlO
psi
2087 0.470
2150 0.487
2112 0.466
2132 0.492
Time
Under
Pressure
(min)
7
10
15
10
10
15
Notes
Blown
OK
Blown
OK
OK
Blown
-------
appear that acceptable panels can be made with a 4% acid content. For
further discussion see the section on Full Scale Production Studies.
Wood Waste-—There were three types of wood waste evaluated. The waste
from Ponderosa Pine, Southern Hard Pine, and Douglas Fir. These wood
wastes were evaluated with different percentages of acid absorption and
with both phosphoric acid and p.-toluene sulfonic acid. The results are
presented in Tables 1-5 through 1-10. As Tables 1-5 and 1-6 show,
Ponderosa Pine does not provide an acceptable product with this system.
Hard Pine, as shown in Table 1-7, provided better results. However, the
best results were obtained with Douglas Fir (see Tables 1-8 through 1-10).
It appears that the best results were achieved consistently with p_-toluene
sulfonic acid. The effect of acid absorption from 0.35% to 4.22% did not
appear to make much difference. Therefore, it was decided to try the full
range of acid absorption. The phosphoric acid required 4.5% to 6.5%
absorption. This percentage with wood waste made the product economically
undesireable. It was hoped that a small wood waste at Washington State
University to save the shipping cost. When work was started at Washington
State University, it was found that the commercially available Douglas
Fir waste had a different geometry than that supplied Material Systems
Corporation by Washington State University in the previous work.
The previous work on 4 inch by 6 inch panels was repeated for the range
of acid absorption. The results are shown in Tables 1-11 and 1-12. Acid
absorption of 0.38% appears to provide properties generally equivalent to
that of 4%. Tables 1-11 and 1-12 used a technical grade of g-toluene
sulfonic acid and Table 1-13 examined a less expensive commercial grade
which resulted in lower properties.
A series of larger panels, 12 x 15 inches, were then made to more closely
approximate full scale production problems with 0.35% acid absorption and
1.78% acid absorption being used. The 0.35% yielded too little resinous
binder and the 1.78% generated too much. Combining the two materials in
quantities of 60% - 67% of 1.78% absorption and 33% - 40% of 0.35%
absorption, yielded potentially acceptable results. These data are shown
in Table 1-14.
Because of the relatively long time required to produce acid treated
particles with the soak-dry method, and the limited time available on the
Washington State University presses, it was decided to use the Washington
State University spray equipment to make acid treated particles.
The acid was sprayed onto the wood chips using 20 grams of 25% p-toluene
sulfonic acid in water, per pound of wood. A small amount of insoluble
impurity (presumably sulfone) had to be filtered out of the solution to
protect the spray nozzles against stoppage. This resulted in a material
with an acid absorption content of 1.05%. The results are shown in Table
1-15.
MOISTURE CONTENT
Peanut Hull Waste-—Only limited variation on the effect of moisture
-------
TABLE 1-5 PROCESSING STUDIES ON PONDEROSA PINE
Acid Solution
Concentration
H PO 7%
3 4
H PO 7%
*J "3
H PO 7%
*J **J
H PO 7%
iJ ~^r
H PO, 7%
3 4
H^PO. 7%
3 4
H-PO 7%
3 - 4
H.PO, 7%
3 4
H PO 7%
O ^
H PO, 7%
3 4
H-PO. 7%
3 4
H.PO,, 7%
3 4
H PO, 7%
3 4
H PO 7%
J ""A
H PO 7%
O **
H3P04 7%
H PO 7%
O *±
H PO 7%
<*j "
H PO 7%
J *±
H PO 7%
«J ^t
H PO 7%
J TC
H PO 7%
J **
H PO 7%
-J ^X
H PO 7%
-> rr
tosic 3% *
tosic 3%
tosic 3%
tosic 3%
500
Process
Conditions
5 min 300°F
5 min 300 F
3 min 325°F
3 min 325°F
5 min 325°F
5 min 325°F
10 min 325°F
10 min 325°F
5 min 350°F
5 min 350°F
3 min 350°F
3 min 350°F
5 min 300 F
5 min 300°F
3 min 325°F
3 min 325°F
5 min 325°F
5 min 325°F
10 min 325°F
10 min 325°F
3 min 350°F
3 min 350 F
5 min 350 F
5 min 350°F
5 min 250°F
5 min 250°F
o +
2 min 295 F +
o
2 min 295 F
psi Pressure
Moisture Shore "D"
Content Hardness
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
6%
6%
6%
6%
6%
6%
6%
6%
6%
6%
6%
6%
7%
7%
7%
7%
Top
61
34
58
32
63
36
65
45
73
47
60
41
48
13
57
9
57
18
75
57
62
35
68
58
55
63
62
57
Bottom
63
37
59
33
68
41
72
51
77
54
76
44
52
17
56
13
62
26
79
59
61
40
71
56
63
66
61
55
Density
1.00
1.00
0.95
0.95
1.05
1.05
1.10
1.07
1.03
1.03
1.07
1.03
0.79
0.83
0.83
0.83
0.83
0.88
1.09
1.09
0.95
0.93
1.06
1.08
0.90
0.90
1.23
0.98
Modulus of
Jlupture
"HJD Soak
699
-
661
-
1103
_
1333
_
1035
_
1242
_
373
-
462
_
776
_
1729
_
1358
_
1758
__
970
__
1726
__
p_-toluene sulfonic acid.
+ Postcured lhr/275°F.
* Postcured lhr/300°F.
10
-------
TABLE 1-6 PROCESSING STUDIES ON PONDEROSA PINE USING MIXED ACID
Process
Conditions
5 min 275 F
5 min 275°F
3 rain 292°F
3 min 292°F
5 min 300°F
5 min 300°F
5 min 310°F
5 min 310°F
4% H_PO
3
Moisture
Content
*
2%
2%
2%
2%
7%
7%
2%
2%
. + 0.7% P-Toluene Sulfonic Acid
4
500
Shore
—
psi Pressure
11 D"
Density
Hardness
Top
63
49
57
44
70
59
68
65
Bottom
65
51
59
46
71
62
62
67
1.01
1.13
0.99
1.13
1.20
1.27
1.10
1.21
Modulus of
Rupture
1134
-
1335
-
1764
-
1396
-
With 7% water content, panel exploded on pressure release.
Panel Blown. Remolded 10 min/50 psi + 10 min/500 psi.
11
-------
TABLE 1-7 PROCESSING VARIABLES, TREATMENT OF HARD PINE WOOD USING PHOSPHORIC
ACID
Moisture Content = 9.9%
Reference
Book
No. 69
MATERIAL
Acid *
(g/lOOg
of chips)
Moisture
+
Mold
Temp
Pressure
psi
MOLDING CONDITIONS
Total Time Blown
Under Area
Pressure (5)
(min)
20-3
20-10
20-2
20-11
20-13
20-1
20-12
20-14
9.42
9.42
6.42
6.42
6.42
3.26
3.26
(50%-9.42)
(50%-3.26)
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
375 500 3.5
320 500 7.0
380 500' 7.0
327 500 5.0
337 500-400 10.0
385 500 10.0
340 500-750 10.0
333 500 7.0
100
60
100
0
80
20
0
0
Reference
Book
No. 69
20-3
20-10
20-2
20-11
20-13
20-1
20-12
20-14
Relative
Shrinkage
PROPERTIES OF MOLDED PANEL
Hardness FLEX-MOR (psi)
Shore D Initial 24-hr water
(Avg of 3) soak (Avg of 2)
Density
Ibs/ft
V. High
Medium
V. High
Medium
Medium
High
Low
Low
65
74
62
61
63
66
57
69
58
63
69
57
66
71
69
77
1480
1530
1170
1195
1130
2055
840
1800
605
585
573
89
694
698
18
362
Given as 100% acid, although 75% acid was used and based on dry wood
chips.
10 grams, 1 hour @ 300 F in forced-air oven.
12
-------
TABLE 1-8 PROCESSING VARIABLES, TREATMENT OF DOUGLAS FIR
Hammermilled to -6+14 mesh; Moisture Content=7.5%
Phosphoric Acid Treatment
MATERIAL
MOLDING CONDITIONS
PROPERTIES OF MOLDED PANEL
Reference
Book
No. 69
20-5
20-7
20-16
20-17
20-18
20-6
20-8
20-15
22-23
22-25
22-26
20-4
20-9
20-19
20-20
20-21
20-22
Acid *
(g/lOOg)
of chips
6.45
6.45
6.45
6.45
6.45
4.58
4.58
4.58
4.58
4.58
4.58
2.33
2.33
2.33
2.33
2.33
2.33
Moisture
(%) +
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
Mold
Temp
(°F)
372
370
358
355
400
372
362
362
372
370
325
378
403
397
358
382
380
Pres-
sure
psi
500
500
500-
750
875
250
500
500
625
750
500
625
500
500
500
875
875
875
Pres-
sure
Relief
Cycle
(min)
None
None
None
None
None
None
None
None
2,3,4
7
None
None
None
None
None
None
3,6
Time
Under
Pres-
sure
(min)
10.0
5.0
10.0
2.5
4.5
10.0
5.0
9.5
4.5
10.0
15.0
10.0
10.0
4.0
10.0
8.0
8.0
Blown
Area
100
0
100
100
0
20
0
0
40
60
40
0
20
0
0
60
0
Rela-
tive
Shrink-
age
High
Med
Med
Low
Low
Med
Low
High
-
-
-
Low
Med
Low
-
-
Density
Ibs/ft
68
67
71
67
-
71
55
68
72
71
58
54
70
54
60
71
69
Hard
ness
Shore
D
77
75
80
72
56
82
59
55
80
65
59
49
74
70
59
74
72
Flex-Mor
(psi)
Initial
(avg of 3)
1545
1780
1565
1420
1400
2345
545
1925
2349
2189
1178
750
2025
703
1007
2020
2018
Given as 100% acid, although 75% acid used and based on dry wood chips.
10 grams, 1 hour @ 300 F in forced-air oven.
-------
TABLE 1-9 PROCESSING VARIABLES, TREATMENT OF DOUGLAS FIR
Hairanermilled to -6+14 mesh; Moisture Content =8.5%
Treated with p-Toluene Sulfonic Acid
MATERIAL
Reference Acid *
Book (g/lOOg
No. 69 of chips)
26-3
26-10
26-1
26-11
26-16
26-19
5.49
5.49
2.21
2.21
0.46
0.46
MOLDING CONDITIONS
Moisture Mold Pressure Pressure Total Time
(%) + Temp psi Relief Under
( F) Cycle Pressure
13.1
13.1
14.3
14.3
4.1
4.1
300
280
340
260
307
392
500
500
500
500
750
750
None 5.
None 5.
None 5,
None 5.
None 15.0
3 7.0
Reference
Book
No. 69
PROPERTIES OF MOLDED PANEL
Blown Relative Density Hardness
Area Shrinkage Ibs/ft Shore D
Flex-Mor psi
Average of 3
26-3
26-10
26-1
26-11
26-16
26-19
100
80
100
0
0
100
V. High
High
V. High
High
-
Medium
70
77
67
63
52
73
50
74
64
72
60
67
1810
2455
1185
1025
V. Weak
2057
* Given as 100% acid, although 75% acid was used and based on dry wood
chips.
+ 10 grams, 1 hour @ 300 F in forced-air oven.
Note: Pine wood shavings from MSC planer (8.5% moisture)
p-Toluene Sulfonic Acid (100%)
14
-------
TABLE 1-10 PROCESSING VARIABLES, TREATMENT OF DOUGLAS FIR
MATERIAL
Reference Acid *
Book g/lOOg
No. 69 of chips
26-4
26-8
26-9
26-12
26-13
26-14
30-3
30-4
26-2
26-5
26-6
26-7
30-1
30-2
30-10
4.22
4.22
4.22
4.22
4.22
4.22
3.99
3.99
1.72
1.72
1.72
1.72
1.72
1.72
1.64
ammermilled to -6+14 mesh; Moisture Content = 6.6%
Treated with p-Toluene Sulfonic Acid
MOLDING CONDITIONS
Moisture
(%) +
9.5
9.5
9.5
9.5
9.5
9.5
7.9
7.9
11.4
11.4
11.4
11.4
11.4
11.4
14.1
Mold
Temp
(°F
275
262
275
272
277
273
345
350
350
307
308
288
353
265
282
Pres-
sure
(psi)
500
500
1000-
500
500
500
500-
45
500-
0
500
1000-
500
1000-
2000
1000-
1000
750
500-
500-
200
Pres-
sure
Relief
Cycle
(min)
None
None
None
None
None
None
None
None
None
1
1,3
1
1,2,3,
4,5,6,
7,8,9
None
None
Time
Under
Pres-
sure
(min)
5
8
5
10
5.5
5
8
6
5
5
10
5
10
10
10
PROPERTIES OF
Blown
Area
(%)
0
0
0,
0
0
0
0
0
100
100
60
0
100
0
0
Rela-
tive
Shrink-
age
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Medium
V. High
High
Medium
Medium
V. High
Medium
Low
MOLDED PANEL
Density
(Ibs/
ft 3)
74
69
67
75
74
75
57
52
54
76
74
65
73
72
56
. . _ _ -i \
Hard-
ness
Shore
D
77
69
66
79
81
84
68
64
44
76
64
45
76
80
48
Flex-
ure
MOR
(psi)
(avg of
3)
2955
1845
1960
3414
3058
2887
1276
862
420
2010
2055
815
744
1351
263
(continued)
-------
TABLE 1-10 (Continued) PROCESSING VARIABLES, TREATMENT OF DOUGLAS FIR
MATERIAL
Reference
Book
No. 69
Acid *
g/lOOg
of chips
MOLDING CONDITIONS
Moisture
Mold Pres-
Temp sure
(°F) (psi)
PROPERTIES OF MOLDED PANEL
Pres- Time Blown
sure Under Area
Relief Pres- (%)
Cycle sure
(min) (min)
Rela- Density
tive (Ibs/
_* i i f. "^ \
30-11
26-15
26-17
26-20
26-21
26-18
1.64
0.35
0.35
0.35
0.35
0.35
14.1
13.4
13.4
13.4
13.4
13.4
303
302
355
407
415
!388
500-
165
750
750
750
750
750
None
5,10
2,5
5,10,11
3
15
15
15
7
15
10
0
0
0
20
60
0
Low
Low
Low
Medium
Medium
Medium
57
56
62
73
71
73
56
55
76
68
73
74
424
234
1264
2386
2701
2391
* Given as 100% acid, although 75% acid was used and based on dry wood chips.
+ 10 grams, 1 hour @ 300°F in forced-air oven.
-------
TABLE 1-11 COMMERCIAL DOUGLAS FIR (-17 MESH) WITH P-TOLUENE SULFONIC ACID
MATERIAL
Reference Acid
Book g/lOOg
No. 69 of
42-41
42-42
42-18
38-14
38-15
4.38
4.38
4.38
4.38
4.38
4 x
6 in. Panels
L MOLDING CONDITIONS
Moisture
ig +
.ips (%
8.
8.
11.
16.
16.
)
2
2
4
4
4
Mold
Temp
(°F)
265
297
265
292
268
Pres-
sure
(psi)
1000
1000
1000
1000
1000
Pres-
sure
Relief
Cycle
(min)
None
None
None
None
1,2,3,
5,8
Time
Under
Pres-
sure
(min)
5
15
5
6.5
5
PROPERTIES
Blown
Area
(%)
0
0
100
80
82
Rela-
tive
Shrink-
age
Medium
Low
Medium
Low
Medium
OF MOLDED PANEL
Density
(Ibs/
ft")
79
45
72
80
82
Hardness
Shore D
82
46
80
77
77
Flex-MOR
(psi)
(avg
3)
3550
274
1749
2210
2940
of
Aldrich Chemical Co. technical grade p-toluene sulfonic acid monohydrate.
10 grams, 1 hour @ 300°F in forced-air oven.
-------
TABLE 1-12 COMMERCIAL DOUGLAS FIR (+17 MESH) WITH P-TOLUENE SULFONIC ACID
OD
4 x 6 in. Panels
MATERIAL
Reference
Book
No. 69
38-16
38-12
42-43
42-40
38-17
42-25
38-13
Acid Moisture
g/lOOg
of chips
0.380
0.380
4.03
4.03
4.03
4.03
4.03
H
(*
7.
11.
7.
7.
10.
10.
16.
u
.)
7
0
3
3
2
2
0
Mold
Temp
(%)
398
395
310
265
268
262
274
MOLDING
Pres-
sure
(psi)
750
750
250
750
750
750
500
CONDITIONS
Pres-
sure
Relief
Cycle
(min)
None
1,2,3,
5,8
None
None
None
None
None
Time
Under
Pres-
sure
(min)
10
10
15
6
5
5
5
PROPERTIES
Blown
Area
(%)
100
80
0
0 '
0
0
0
Rela-
tive
Shrink-
age
Medium
Low
Medium
Low
Medium
High
Medium
OF MOLD:
Density
dbs/
ft )
76
70
52
77
78
70
74
Hardness
Shore D
79
76
70
77
80
80
67
Flex-
MOR
(psi)
(avg of
3)
2588
2290
1104
3403
3016
2414
1450
Aldrich Chemical Co.. technical grade p-toluene sulfonic acid monohydrate.
10 grams, 1 hour @ 300°P in forced-air oven.
-------
TABLE 1-13 COMMERCIAL DOUGLAS FIR WITH ULTRA TX ACID
4 x 6 in. Panels
MATERIAL
Reference Acid Moisture
Book (g/lOOg + (%)
No. 69 of chips
MOLDING CONDITIONS
Mold Pressure Pressure Total Time
Temp (psi) Relief Under Pressure
( F) Cycle (min)
42-32
42-33
42-34
42-29
42-30
42-31
0.305
0.305
0.305
0.305
2.384
2.699
10.0
10.0
10.0
10.0
13.3
13.0
390
370
375
265
260
270
750
750
750
500
500
500
7
7
7,14
None
None
None
12
12
20
5
5
5
Reference
Book
No. 69
PROPERTIES OF MOLDED PANEL
Blown Relative
Area Shrinkage
Density Hardness
(Ibs/ft ) Shore D
Flex-MOR
(psi)
(avg . of
3)
Avg.
Particle
Size
42-32
42-33
42-34
42-29
42-30
42-31
0
100
100
0
0
0
High
High
High
Medium
Medium
70
72
72
48
62
65
78
78
73
67
77
1375
1673
1813
1477
2094
On 17 mesh
On 17 mesh
On 17 mesh
On 17 mesh
Thru 17
Witco Chemical Co. commercial grade p_-toluene sulfonic acid.
"*" 10 grams, 1 hour @ 3008F in forced-air oven.
19
-------
TABLE 1-14 LONGVIEW DOUGLAS FIR (ON 20 MESH)/P-TOLUENE SULFONIC ACID
12" x 15" x 0.5"
MATERIAL
Reference Acid Moisture
Book (g/100g WSU
No. 69 of chips) Method S
MOLDING CONDITIONS
Flatten Pressure Pressure Time Under
Temp psi Relief Pressure
( F) Cycle (min)
(min)
46-10
46-11
46-12
46-13
46-14
46-20
46-22
46-15
46-16
46-17
46-18
46-19
46-21
0.35
0.35
0.35
0.35
1.78
1.78
1.78
75%-1.78
25%-0.35
75%-1.78
25%-0.35
67%-1.78
33%-0.35
67%-1.78
33%-0.35
67%-1.78
33%-0.35
60%-1.78
40%-0.35
1.5
1.5
1.5
5.5
3.0
3.0
3.0
3.6
3.6
3.8
3.8
3.8
4.0
390
390
390
390
300
300
300
310
310
300
300
300
300
750-1500
1100-1500
1100
1100
1000
1000
1000
1000
750
750
1000
1000
1100
7
7
7
7
None
None
None
None
None
14 blew
None
None
None
12
12
12
12
5
7
7
6
8
15
8
10
7
(continued)
20
-------
TABLE 1-14 (Continued) LONGVIEW DOUGLAS FIR (ON 20 MESH)/P-TOLUENE SULFONIC ACID
PROPERTIES OF PANEL
Reference
Book
No. 69
46-10
46-11
46-12
46-13
46-14
46-20
46-22
46-15
46-16
46-17
46^-18
46-19
46-21
Time to
Relieve
Pressure
(min)
0
0
0
0
0
1
0
0
1
3
2
2
0
Charge
Size
(grams)
1800
1800
1600
1600
1700
1800
1800
1700
1700
1900
1800
1800
1800
Thickness
(0.5)
-
-
-
0.470
-
-
0.505
0.502
-
0.491
0.504
0.508
Density
Ibs/ft
_
-
-
61
70
-
-
65
64
-
70
68
68
Internal
Bond
IB/psi
_
-
-
-
-
-
-
-
-
-
-
22
14
Flexure Notes
MOR
psi
_
-
-
-
-
-
-
-
-
-
-
1516
2096
MOE
xlO6
psi
Blown
Blown
Blown
Crumbly
Blown
Blown
Blown
Blown
- Blown
Blown
Blown
0 . 594 OK
0 . 783Blown
21
-------
TABLE 1-15 LONGVIEW DOUGLAS FIR (+20 MESH)/P-TOLUENE SULFONIC ACID SPRAY
METHOD
12" x 15" x 0.5"
MATERIAL
Reference Acid Moisture Flatten
"Book (g/lOOg WSU Temp
No. 61 of chips) Method % (°F)
MOLDING CONDITIONS
Pressure Pressure Time
psi Relief Under
Cycle Pressure
(min) (min)
50-40
50-41
50-42
50-43
1.046
1.046
1.046
1.046
7.8
7.8
7.8
7.8
325
325
340
340
1100
1100
1100
1100
None
5
5
6
8
10
10
10
PROPERTIES OF
Reference
Book
No. 69
50-40
50-41
50-42
50-43
Time to
Relieve
Pressure
(min)
0
1
1
1
Charge
Size
(grams)
1800
1800
1800
1800
Thickness
0.519
-
0.492
0.512
PANEL
Density
Ibs/ft
68
_
68
0
65
Flexure Notes
MOR MOE
psi xlO
psi
Blown
- - OK
2402 0.781 Blown
"2144 0.685 Blown
22
-------
content was evaluated for peanut hull waste. This study is reflected in
Table 1-4. No real effect is discernible. For the purpose of the large
scale work a moisture' content between 5% and 6% was selected. This was
similar to that now used in the industry and was considered on the basis
of the data from Table 1-4 to be the most reliable.
Wood Waste—The results from limited studies on the effect of moisture
on wood waste are presented in Tables 1-11 through 1-13 for 4 inch by 6
inch panels. These results show no real trend. However, when tests were
conducted on larger panels (12 inch by 15 inch) it was shown that
increasing the water content from 1.5% to 5.5% assisted the processing of
0.35% acid absorbed mixture (see Table 1-14). Further studies showed that
a mixture of 0.35% and 1.78% acid absorbed material with 3.8% moisture was
better (see Table 1-14). Table 1-15 presents the results of a study on
an acid content similar to the mix in Table 1-14 but with a 7.8% moisture
content and achieves similar results. This indicates that moisture content
has an effect on processing although of less importance than the other
parameters.
TIME UNDER PRESSURE
Peanut Hull Waste—Pressure and time are interrelated. The initial studies
with time under pressure included the effects of water soak. Table 1-16
presents the effect of molding time on properties for a 375 and a 500 psi
cure with a material moisture content of 13%. Table 1-17 presents similar
data for a material moisture content of 9%. There is a definite relation-
ship between press cure time and density and a relationship between density
and strength retention with water pick up. Longer press cure times can be
expected to produce more resin. Increased resin decreases voids which re-
duces water pick up that effects strength loss. There is a point where
further generation of resin becomes self-defeating because the chemical
reactions forming the resinous material will eventually lead to excessive
shrinkage and eventual cracking of the laminate. From this study, it can
be concluded that a minimum of 5 minutes press times is required at a cure
temperature of 375°F to generate sufficient resin for satisfactory
properties.
If the generation of resin is a function of time and temperature, it can
be theorized that shorter times with higher temperatures will produce equal
results to longer times and lower temperatures. Temperatures up to 450 F
were evaluated with shorter times. The results are presented in Table 1-18.
As can be seen, times less than 5 minutes show a marked decrease in initial
properties as well as resistance to water.
Wood Waste—The work on peanut hulls was applicable to wood waste. The
effects of variations of time under pressure for Ponderosa Pine are
presented in Table 1-5, for Hard Pine in Table 1-7, and for Douglas Fir in
Tables 1-8 through 1-15. Cure times of 5 minutes appear to be minimum
and between 5 and 10 minutes optimum.
23
-------
TABLE 1-16 EFFECT OF MOLDING TIME ON PROPERTIES
Peanut Hull Composites
Cure Temp - 375 F; Cure Pressure - 500 psi; Moisture Content - 13.0%
Specimen No.
Mold Time
(mins)
Weight, grams
Hrs, HO soaked
0 22 24
Plexural Strength, psi
Hrs, HO soaked
0 24
1-1
1-2
1-3
Average
6-1
6-2
6-3
Average
7-1
7-2
7-3
Average
8-1
8-2
8-3
Average
9-2
9-1
9-3
Average
Whittaker *1
2
Average
Average
4
5
6
15
15
15
15
15
15
10
10
10
10
10
10
5
5
5
6
6
6
6
6
6
25.7
28.6
27.7
26.4 27.7 38.9
26.1 27.1 28.7
25.0 26.4 28.5
27.7
26.4
27.4
28.4
26.5
29.4
27.5
29.4
28.5
62.5
58.9
60.5
30.8
29.6
31.7
__
39.9
41.4
_
-
_
34.3
32.6
36.2
_
42.6
43.4
_
-
_
1268
1443
1215
1308
60.2 88.1 91.7
61.3 87.4 91.7
60.9 92.4 96.5
1765
1647
1561
1658
1078
358
333
366
352
933
674
576
728
170
100
135
44
66
29
46
Panel made at Whittaker Corp. on 24" x 24" press to 0.5" thickness under
Phase II. All other specimens made at MSC to 0.250-0.300" thickness.
Moisture content of peanut hulls used to make panel at Whittaker Corp.
was 14.0%.
24
-------
TABLE 1-17 EFFECT OF MOLDING TIME ON PROPERTIES
Specimen No.
2-1
2-2
2-3
Average
3-1
3-2
3-3
Average
1-1
1-2
1-3
Average
6-1
6-2
6-3
Average
5-1
5-2
5-3
Average
4-1
4-2
4-3
Average
8-1
8-2
8-3
Average
Average
11-1
11-2
11-3
Average
12-1
12-2
12-3
Average
* _
Peanut Hull
o
75 F; Cure Pressure
Composites
- 500 psi; Moisture Content - 9.0%
Mold Time* Weight, grams Weight, grams Flexural Strength,
(mins) Hrs, Hot
0
5
5
5
-
5 30.6
5 29.7
5 29.9
-
5 -
5
5
-
3
3
3
-
3 30.4
3 31.2
3 30.4
- _
3
3
3
-
2
2
2
_ _
2
2
2 _
-
1
1
1
_ _
1 31.1
1 30.3
1 31.6
_ M
rlorl fnr 1 and 2 minu
Box Hrs, HO Soak psi Hrs Exposure
24 2 24 0 24HB
- - 1164 -
- - 1112 -
- - 1220 -
- - 1165 -
34.2 - - - 283
33.2 - - - 306
33.5 - - 232
- - - 274
29.6 43.7
30.0 41.8
28.9 41.9
_ -
- - 1293 -
- - 1194 -
- - 1240 -
- - 1242 -
32.9 _ - 179
33.4 - - - 304
32.1 - - 177
_ - - 220
31..6 55.6
31.7 50.7
31.5 52.7
_ -
- - 407 -
- - - 342 -
- - 288 -
- - - 346 -
- - 37
- - - - 64
- - 43
- - 48
- - - 284
- - - 333 -
- - - 352 -
- - - 323 -
34.1 - - ~^~ 11
32.5 - - - 12
33.9 - - - 26
- 16
ites were useless after 2-hr. soak.
24H2°
_
_
_
_
_
_
_
-
178
196
200
191
_
_
_
-
_
_
—
_
43
52
62
52
_
_
-
-
.—
-
-
-
-
-
-
-
-
-
-
_
25
-------
TABLE 1-18 EFFECT OF HIGH PRESS TEMPERATURES AND SHORTER PRESS TIMES ON
Press
Temp
°F
410
410
410
425
425
425
450
410
410
425
425
410
410
425
425
425
425
450
450
450
450
450
450
450
PROPERTIES OF PEANUT
Pressure
Press Time
(minutes)
5
5
5
5
5
5
5
3
3
3
3
2
2
2
2
1
1
1
1
1
1
1
1
1
- 500 psi
HULL WASTE PANELS
Moisture Content
% Weight Gain
after Water
2hrs
10
11
9
12
11
17
All attempts
blowing.
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
8
Soak
24hrs
40
35
18
14
20
44
to make
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
22
Flexural
- 3%
Strength
, psi
after Water Soak
0 hrs
1672
1600
1502
1403
1764
1330
2 hrs
1279
1263
1100
1184
813
826
panels resulted in
670
537
937
1303
504
596
1104
1117
380
307
740
809
285
224
989
85
-
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
-
24 hrs
191
274
988
1052
883
184
serious
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
63
These specimens fell apart after 2-hr water soak.
26
-------
PRESS TEMPERATURE
Peanut Hull Waste*—The effects of varying temperatures on the material
properties are presented in Tables 1-4, 1-16, and 1-18. The possibility
of a panel blosing increases with temperature. The temperature range that
consistently provided dense, structurally sound panels was 350°F to 400°F.
The theory that increased temperature reduces press time was proven inef-
fective in the section Time Under Pressure. Therefore, 375°F was chosen
as the starting point for full scale studies.
Wood Waste—The fact that wood generates more volatiles than peanut hulls
results in more possibilities for blowing. Therefore, lower temperatures
were evaluated. Tables 1-5 and 1-6 present data for Ponderosa Pine, Table
1-7 for Hard Pine and Tables 1-9 through 1-15 for Douglas Fir. Tempera-
tures from 250 F to 400 F were evaluated. The temperature range that
consistently provided dense and structurally sound panels was 300°F to
350°F.
PRESSURE TECHNIQUE
Peanut Hull Waste—For the purpose of this discussion a pressure relief
cycle is defined as: rapid release, at a predetermined time, of the pressure
applied to the material. The press plattens are slightly opened to allow
the vapor build up in the material to be released. The pressure is then
reapplied and maintained until the press is completely closed.
"Decompression -time" is a gradual reduction of the final pressure to 0 psi
over a predetermined time after the press time selected was complete.
Table 1-4 presents a summary of the effects on the properties of 4 inch by
6 inch panels with and without pressure relief cycles. For this size panel
and for pressure over 500 psi, the pressure relief cycle improves the
properties of the product. This is particularly true where multiple cycles
are used. However, the fabrication of the larger panels showed less
tendency for blowing without pressure relief (see Tables 1-18 and 1-30 in
the section of Full Size Panel Production). This table also presents the
effect of "Decompression Time". It appears that a decompression of slight-
ly over one minute will materially improve the product.
Wood Waste—The wood waste small panels, Tables 1-7 through 1-15, show that
pressure relief cycles have less effect than that of peanut hulls. The
fabrication of the larger panels (see Table 1-33 in the section on Full
Scale Panel Production) showed that one pressure relief cycle appears to
reduce blowing considerably especially when the release occurs two thirds
through the press time. As in peanut hull waste, a decompression time of
slightly over one minute will improve the product.
EFFECTS OF POST CURE
The post cure studies were conducted on peanut hull waste. The primary
purpose was to determine if the properties of short press time products
could be improved. The first effort was an attempt to improve the
properties of a 1 minute cure at 450°F. The results are presented in
Table 1-19.
27
-------
TABLE 1-19 EFFECT OF POST CURING CONDITIONS ON PEANUT SHELL PROPERTIES
Press molded 1 minute at 450 F
Specimen
Post Cure Conditions
Wt.,grams
hrs, HO soak
0 24
Flexural Strength
hrs, HO soak
0 24
26-2
27-1
28-1
28-2
28-3
1 hr/275^F
2 hr/275°F
16hr/300°F
16hr/300°F
16hr/300°F
*
*
*
31.9
30.6
* 853
* 659
* 468
50.6
49.0
113
101
Specimens failed during 2-hr. soak.
28
-------
There was some improvement in flexural strength compared to those
without post cure (Table 1-18) but nothing considered worthwhile. However
the resistance to water was improved. It was then determined to examine
the effects of post cure on a more conventional processing procedure.
Test panels were fabricated under identical conditions but subjected to
various post cure and pre-cure conditions, such as using peanut hulls
having two different moisture contents. The pressing conditions of 5
minutes, 500 psi at 375 F were selected because there was a sharper
increase in strength beyond 5 minutes which potentially made post cure
effects more noticeable, either better or worse. Table 1-20 gives the
data from small panels to show changes in dimensions, weight and density
for various post cures.
Post cure of the panels indicated definite shrinkage in dimensions, hence
volume, weight and density changes. Thickness was the most sensitive to
change of the dimensions with weight exhibiting greater changes than vol-
ume or density. There seemed to be no correlation in these data. The
weight loss is important because this is now considerably greater than the
moisture content of the treated hulls prior to processing. This is
indicative that water is evolved by a dehydration reaction, probably from
a polymerization reaction during pressing. All six panels were next cut
into flexural specimens. Five were used for water soaking with one serv-
ing as control. Results are shown in Table 1-21. It is seen that the
post cure resulted in a definite decrease in percentage weight, volume and
density gained on water exposure. The percentage flexural strength loss
after 24 hour water immersion was decreased under all post curing condi-
tion. Strength increases of over 200% are not realistic and are probably
due to a faulty control specimen providing a low flexural strength level.
The above procedure was repeated except that the panels were press cured
for only 3 minutes at 375°F and then post cured (see Tables 1-22 and 1-23).
The same effects were found again with post curing causing further decrease
in weight, dimensions and density. It appeared that the 3 minute cure
resulted in smaller decreases in volume than panels cured for 5 minutes,
both were cured at 375°F, while the change in density was larger for the
3 minute cure. This phenomenon is reasonable only if weigh changes are
about the same for either cure and this appears to be the case. Overall,
however, there appears to be no improvement in properties that would
justify a post cure in a production process.
EFFECTS OF PRE-HEAT
This study was conducted with peanut hull waste only. There are several
important reasons why the use of preheated, treated peanut hulls is of
interest:
a) They can go directly from drying ovens to the press and this would
save energy, and permit faster heat-up to desired temperature.
29
-------
TABLE 1-20 EFFECT OF POSTCUKE ON PEANUT HULL PANELS
Moisture Content = 9%; Processed - 5 mins @ 500 psi @ 375 F
Postcure Conditions
None (Control)
30 mins/400°F
2 hrs/375°F
16 hrs/325°F
2 hrs /375° + lhr/425
16 hrs/3250 + Ihr/
375°
No . Dimensions , cm
W X T x L
1 10.2 0.81 15.2
2 10.2 0.85 15.3
3 10.2 0.82 15.2
4 10.2 0.92 15.2
5 10.2 0.84 15.2
6 10.2 0.74 15.2
2 10.1 0.76 15.0
3 10.0 0.81 15.0
4 10.1 0.88 15.1
5 9.8 0.76 14.7
6 9.9 0.69 14.8
TABLE 1-21 EFFECT OF POSTCURE ON
Volume Weight Density
cm grams g/cm
125.1 116.9 0.96
132.7 116.6 0.88
127.1 115.0 0.90
142.6 117.5 0.82
130.2 117.7 0.90
114.7 120.0 1.05
115.1 98.2
121.5 94.5
134.2 105.5
109.5 89.3
101.1 97.3
0.85
0.78
0.79
0.82
0.96
PEANUT HULL PANELS
Moisture Content = 9%; Processed - 5 mins @ 500 psi @
Postcure Conditions
None
30 mins/400°F
2 hrs/375°F
16 hrs/325°F
2 hr/375° + hr/425°
f\ f*\
16 hr/325 + 1 hr/375
% Weight % Volume
Gain, Hrs Gain, Hrs
Water Soak Water Soak
2 24 2 24
28.6 50.0 7.8 23.2
6.6 13.7 2.3 4.8
6.3 11.9 3.1 6.8
6.7 20.1 6.1 13.6
10.1 21.6 10.0 10.6
6.0 13.8 7.0 6.5
% Density
Gain , Hrs
Water Soak
2 24
19.8 20.
3.7 9.
3.9 4.
1.2 6.
0 9.
0 6.
375°F
Flexural
Strength, psi
Water Soak , Hrs
0 2 24
9 748 521 114
2 421 938 1030
7 1254 691 1071
4 1239 882 318
2 901 176 629
4 1149 927 908
30
-------
TABLE 1-22 EFFECT OF POSTCURE ON PEANUT HULL PANELS
-o.
Postcure Conditions No
None (Control)
30 mins/400 F
2 hr/375°F
15 hr/325°F
2 hr/375° + lhr/425°
16 hr/325° + Ihr/
375°
t = 9%; Processed - 3 mins @ 500 psi @ 375'
Dimensions, cm Volume Weight
W
10.
10.
10.
10.
10.
10.
10.
10.
10.
9.
x T x
2
2
2
2
3
2
1
0
0
7
0.
0.
0.
0.
1.
0.
0.
0.
0.
0.
76
74
74
73
04
90
71
73
71
93
15
15
15
15
15
15
15
15
14
14
L
.2
.2
.2
.2
.3
.3
.0
.0
.9
.5
cm
117
114
114
113
163
140
107
109
105
130
grams
.8
.7
.7
.2
.8
.5
.6
.5
.8
.8
120
121
119
119
127
126
102
101
96
84
.8
.0
.3
.9
.5
.5
.2
.5
.0
.4
10.1
0.85 15.0 128.8
104.9
Density
g/cm
1.03
1.05
1.04
1.06
0.78
0.90
0.95
0.93
0.91
0.65
0.81
TABLE 1-23 EFFECT OF POSTCURE ON PEANUT HULL PANELS
Moisture Content = 9%; Processed - 3 mins @ 500 psi @ 375°F
Postcure Conditions
% Weight
Gain ,. Hr.
Water Soak
2 24
% Volume
Gain, Hr.
Water Soak
2 24
% Density
Gain , Hr
Water Soak
2 24
Flexural
Strength,
Water Soak , hr
0 2 24
None
30 mins/400°F
2 hr/375°F
16 hr/325°F
2hr/375° + lhr/425°
16/325° + lhr/375°
19.1
8.0
7.6
6.5
9.9
7.4
51.6
19.6
23.1
18.1
33.6
29.2
4.5
3.7
0
1.9
4.6
1.6
28.7
5.7
8.4
5.9
8.7
8.1
13.8
3.4
8.1
4.5
6.6
4.8
9.5
12.8
13.7
11.8
23.9
20.3
398 69 26
1117 1102 589
988 494 561
1067 721 519
322 236 259
650 578 192
31
-------
b) They can be used to pre-advance the resin to remove a part of the
condensation volatiles, thus minimizing blowing tendencies.
c) Achieve more rapid resin flow to improve binder action, thus also
minimizing blowing.
After treatment with acid, the peanut hulls were dried to 9% and 3%
moisture content and immediately put into sealed containers. Panels were
prepared by taking sufficient, 130 grams of peanut hulls, preheating 15
minutes at 400°F, immediately transferring to a preheated mold and press-
ing.
Using peanut hulls with 9%. moisture, data are presented in Tables 1-24 and
1-25 for various press schedules on material preheated 15 minutes at 400 F.
It was found that after preheat treatment, the moisture content was in the
range of 6.5 - 8.0% water.
The most interesting data are seen in Table 1-25. With press times
decreasing, changes in volume and weight vary inversely while strength
varies directly with press time regardless of water soaked or not. The
above mentioned tests were repeated but using treated peanut hulls having
a 3% moisture content. Data are given in Tables 1-26 and 1-27. Again
these same trends were found. The best overall performance resulted with
the 15 minute cure on material containing 3% moisture before preheat. The
data also verified previous results that indicated a minimum press time
of 5 minutes at 375 F. Overall, however, the use of a preheat to improve
a particleboard made from peanut hulls is apparently effective with longer
press cures and is not attractive commercially except for use in very humid
or damp environments.
PARTICLE SIZE
Peanut Hull Waste—The effects of particle size became of interest after
the first large scale production. Evidence indicated that a high concen-
tration of fine particles can cause blowing because of the following:
a) Fine particles block the escape route of water vapor and volatiles
tiles between larger particles.
b) Fine particles can react more rapidly and produce more resinous binder
in situ and this can then undergo reaction to a higher degree for a
given time period, than can coarse particles.
In blown panels, especially those that exploded, it was noticed that dark
areas were formed away from edges, where higher degrees of resin formula-
tion had taken place. These dark areas were always found where panels
exploded and where there were always considerably more fine particles
present. Consistent with this is a cracking phenomenon. Such behavior
is consistent with formulation of excess resin ans subsequently more
extensive reaction giving off more water (increasing internal pressure)
and also causing excessive shrinkage leading to cracking. Shrinkage is
rarely seen in 4 inch by 6 inch panels.
32
-------
TABLE 1-24 USE OF PREHEATED PEANUT HULLS MOLDED AT VARIOUS PRESS CYCLES
Moisture Contact of Treated Material =
Preheat Cycle = 15 min @ 400 F
Press Cycle
Weight
Before
Preheat
Weight
After
Preheat
Weight
After
Pressing
Density
g/cm
15 min/375 F
10 min/375°F
5 irdn/375°F
2 min/375°F
1 min/375°F
130.0
130.0
130.0
130.0
130.0
126.6
127.4
127.7
128.6
128.0
114.2
118.7
113.3
120.0
119.8
1.15
1.07
1.08
0.95
0.96
TABLE 1-25 USE OF PREHEATED PEANUT HULLS MOLDED AT VARIOUS PRESS CYCLES
Effect of Water Exposure on Properties
Press Cycle
% Weight Change
Water Soak
2 hr 24 hr
% Density Change
Water Soak
2hr 24 hr-
Flexural Strength
psi Water Soak
Ohr. 2hr, 24hr
15 rains/375 F 3.8
10 mins/375°F 5.3
5 mins/375°F 11.8
2 mins/375°F 48.8
1 min/375°F 33.4
7.4
17.4
39.4
66.7
51.1
0
2.8
5.8
23.6
15.3
+2.7 2043 2230 1484
+12.1 1894 1109 458
+14.3 1745 585 76
-3.5 762 27 0
-1.0 578 34 0
33
-------
TABLE 1-26 USE OF PREHEATED PEANUT HULLS MOLDED AT VARIOUS PRESS CYCLES
Moisture Content of Treated Material - 3%; Preheat Cycle = 15 mins @ 400 F
Press Cycle
Weight g
Before
Preheat
Weight g
After
Preheat
Weight g
After
Pressing
Density
g/cm
15 mins/375 F 130.0
10 mins/375°F 130.0
5 mins/375°F 130.0
2 mins/375°F 130.0
1 min/375°F 120.0
128.0
125.5
128.0
120.4
126.4
108.2
111.1
113.1
116.2
118.4
1.08
1.11
1.00
0.81
0.81
TABLE 1-27 USE OF PREHEATED PEANUT HULLS MOLDED AT VARIOUS PRESS CYCLES
Effect of Water Exposure on Properties
Moisture Content of Treated Material = 3%; Preheated Cycle = 15 min @ 400°F
Press Cycle
% Weight Change
Water Soak
24 hrs 24 hrs
% Density Change
Water Soak
2 hrs 24hrs
Flexural Strength,
psi Water Soak
Ohrs 2hrs 24hrs
15 mins/375 F 4.1 8.3 1.8
10 mins/375°F 4.6 10.6 1.9
5 mins/375°F 11.1 16.1 6.9
2 mins/375°F 31.1 57.0 18.0
1 min/375°F 65.2 63.7 25.6
0.9
8.4
12.3
16.1
29.3
2080 1257 1211
2059 1632 816
929 1422 1311
1106 126 44
613
0
43
34
-------
For all future work it was decided to remove any material that would
pass through a -32 to -42 Taylor Screen. Also it was decided to mix
all material thoroughly in order to insure a random particle size.
Wood Waste—The initial work on Douglas Fir was on a hammermilled material
where the particle sizes ranged from -6 to 14 mesh. Results are presented
in Tables 1-8 through 1-10. The final work was done with commercial wood
waste. Table 1-11 presents results with particles less than 17 mesh and
Table 1-12 presents results with particles greater than 17 mesh. As with
the peanut hull waste, the large particles provided a better product.
FULL SIZE PANEL PRODUCTION
Peanut Hull Waste—Panels were nominally fabricated at the Washington State
University's facility, 50 inch by 50 inch by 0.50 inch thick and 12 inch
by 15 inch by 0.50 inch thick. The 50 inch by 50 inch panels were molded
in an oil heated press and the 12 inch by 15 inch panels in an electri-
cally heated press. Stops were used to control final board thickness.
The mat was formed by hand in a deckle box, Figure 1-1, on a sheet of
Mylar placed on an aluminum caul plate. A heavy board was placed on the
mat in the deckle box with the frame and board removed (Figure 1-2) . The
mat was covered on top by Mylar film, Figure 1-3, and an aluminum caul
plate, Figure 1-4. The entire assembly, except for the particle board
base, was slid into the press (Figure 1-5) and pressure applied (Figure
1-6). For the small panels, a single steel caul plate was used in the
bottom of the mat without Mylar film. A steel wear plate mounted on the
bottom of the top press plate was in contact with the top of the mat
during pressing. No sticking was encountered without Mylar.
All boards, except one, that remained intact were tested for modulus of
rupture (MOR) which is also called flexural strength, and internal bonding
(IB) according to ASTM-D-1037 "Standard Methods of Evaluating the Proper-
ties of Wood Base Fiber and Particle Panel Materials". Some modulus of
elasticity (MOE) were also determined. Figure 1-7 shows the test specimen
cut up pattern of the intact panels evaluated. Portions of some of the
"blown" panels were also treated. The results of the pressing operations
are given in Tables 1-28A and B for large panels.
The processing variables study, described in the previous text, gave
excellent indication of problems and approximate processing requirements.
This work was conducted with small (4 inch by 6 inch) specimens. Fabri-
cation of panels whose areas and volumes were 78 and 312 times greater,
respectively, needed some changes in moisture content. Rather than use
moisture contents of 7% to 13% the first material was dried to about 6%
as used for wood particles in the particle board industry. Temperatures,
pressures and press times were as determined on small panels except that
a decompression time of 3 minutes was used for the large panels.
A wide variation in physical and mechanical properties was found. Three
panels were good and eight exploded during decompression, all boards had
35
-------
Figure 1-1. Forming mat,
Figure 1-2. Placing board on mat,
-------
Figure 1-3. Covering with Mylar.
Figure 1-4. Placement of aluminum caul plate.
-------
Figure 1-5. Insertion into press.
Figure 1-6. Molding.
-------
24"
TUB "M » T. ,
Bd. #
MOR/MOE & IB 4
MOR/MOE & IB 3
MOR/MOE S IB 2
MOR/MOE & IB 1
4"
2"
4"
2"
4"
2"
4"
2"
48'
f
48"
Figure 1-7. Test specimen cut-up diagram. 50 BY 50 INCH PANELS,
39
-------
Bd. #
j
MOR/
MOE
IB
IB
IB
IB
MOR/
MOE
^2" — j-»- 2"—
_•-•*. 1 1 "
i
2"
13
II
Figure 1-7. (Continued) Test specimen cut-up diagram. 12 BY 15 INCH PANELS.
40
-------
TABLE 1-28A PRESSING CONDITIONS AND AVERAGING PHYSICAL PROPERTIES OF
THE 50 BY 50 IN. PANELS
Board No. Furnish Weight Press Press Press Comments
MC (%) Furnish Temp Time Pressure
Used, Ibs ( F) (min) psi
(B) 1 5.5 60 375 5+3 * 500 Did not reach
stops, good
board
2 8.5 51 375 5+3 500 Reached stops
board blew @
loops
3 4.8 51 375 5+3 500 Reached stops
good board
(A) 4 5.9 51 375 5+3 500 Board Blew
5 4.6 47 375 5+3 500 Board blew at
100 psi
6 4.6 47 + 375 7+3 500 Reached stops,
board blew
7 4.2 47 375 5+3 700 Reached stops,
board blew @
40 psi
(B) 8 +" 5.5 47 375 5+3 500 Board blew
(blistered)
9 5.7 42 * 375 5+3 500 Reached stops,
board blew
10 6.8 42 375 5+3 500 Reached stops,
board blew
(A) 11 5.9 42 375 5+3 500 Reached stops,
good board
Five minutes under pressure, three minutes for decompression.
Thermocouple study. Put down 23.5 Ib hulls, put in Thermocouples and
and added remaining 23.5 Ib of hulls.
* Five minutes under pressure, three minutes for decompression.
* Thirty-two material screened out.
41
-------
TABLE 1-28B PRESSING CONDITIONS AND AVERAGING PHYSICAL PROPERTIES OF THE
50 BY 50 IN. PANELS
Board No. Internal Bond Modulus of Rupture and Modulus of Elasticity
SG PSI SG MOR (psi) MOE (psi x 10 )
tB) 1 0.96 49 1.02 710
(A)
(B)
9 0.89 14 -
0.236
(A) 11 0.88 101 0.89 940 0.567
2
3
4
5
6
7
*
8
9
10
11
1.06
0.91
1.10
-
-
_
0.98
0.89
0.87
0.88
193
146
-
-
-
„
128
14
103
101
1.03
-
-
-
-
_
0.96
-
0.86
0.89
1550
-
-
-
-
_
1260
-
900
940
Five minutes under pressure, 3 minutes for decompression.
42
-------
blisters or severe shrinkage cracks. Portions of the blown panels, such
as in No. 2, that remained intact had rather high internal bond values
and the best modulus of rupture value of all the large panels. All the
blowing problems occured during pressure release after the press cycle.
The blowing phenomenon is caused by excessive vapor pressure in the panel
during the press cycle. This vapor pressure emanates from water and/or
other vaoltile materials. Water can come from that already present in the
peanut hulls after processing or from the resinous binder being formed
in situ from pentoses. This involves a cyclodehydration reaction in which
36% of the pentose molecule is evolved as water. This is a large amount
to be given off and occurs during the press cycle under elevated temper-
atures. To give some idea of potential internal vapor pressure (due to
water) during press cure, the following table was taken from the Handbook
of Chemistry and Physics, Steam Tables:
Temg. F Pressure, psi
200 11.5
250 29.8
300 67.0
350 134.6
375 184.3
400 247.2
To determine relative stream pressures generated during cure, thermocouples
were embedded into the mat and temperature recorded during cure. The
resulting temperatures are shown in Figure 1-8. It is obvious that vapor
pressure within the board easily exceeds that of the internal bond strength
of the panels. This is particulary so in a hot panel and even more so in
this study where the resin matrix is thermoplastic at this temperature.
Commerical particle board is at high and low moisture contents in the hulls.
Blowing with a high (8.5%) moisture content is understandable but blowing
with a 4.2% moisture content is unusual to say the least. One explana-
tion is that low moisture content made for more difficult resin flow due
to insufficient plasticizing action. However, evidence indicates that
particle size causes blowing also because of the following:
a) Fine particles block the escape route of water vapor and volatiles
between larger particles.
b) Fine particles can react more rapidly and produce more binder in situ.
This binder can then undergo reaction to a higher degree for a given
time period than can coarse particles.
In the blown panels, especially those that exploded, it was noticed that
dark areas were formed away from edges, where higher degrees of resin
formation had taken place. These dark areas were always found where panels
exploded and where there were always considerably more fine particles pre-
sent. Consisten with this is a cracking phenomenon clearly seen in Figure
1-9. The crack is 0.75 inches wide and note the dark area around the
crack. Such behavior is consistent with formation of excess resin and
subsequently more extensive reaction giving off more water (increasing
43
-------
Start of Decompression
400.
s
-------
Figure 1-9. Cracking phenomenon.
Figure 1-10. Steam generation.
-------
internal pressure) and also causing excessive shrinkage leading to crack-
ing. This meant that another, and unknown variable, was operative during
the first 9 panels and could be the reason why panel No. 8 was unsuccess-
ful even though it was a duplicate of No. 1 in all other respects. The
processed material, as received by Washington State University, was not
mixed or randomized as is normal procedure. Consequently there could be
major variations in particle size distribution in the various drums. Once
this parameter appeared to be a factor, examination of individual drums
revealed large differences in the amount of fines in various drums.
In the next dull size study, the entire material was pre-screened to remove
the -32 to -42 material. Figures 1-10 and 1-11 show steam evolution short-
ly after the press cycle starts and near the end of the decompression cycle,
where blowing occurs. Figures 1-12 and 1-13 show removal of a successful
panel from the press. Figure 1-14 shows a 50 inch by 50 inch panel cooling
anf Figure 1-15 is that of a 24 inch by 24 inch specimen cut from the
larger panel. Figure 1-16 shows where a panel exploded while in the press.
Figure 1-17 was taken from a wall 40 feet distance from the press, showing
debris from an explosion while Figure 1-18 shows one segment of panel hurled
about 30 feet from the press.
The curves in Figure 1-8 are interesting for other reasons. They show that
the peripheral regions of the panel never exceeded 300 F at the end of the
press cycle. Since volatiles can escape more readily near edges of panels,
the cooling effect is greater and the absence of insulation around the panel
permits greater heat loss to the environment. The shape of curve 3 shows
no sudden drop in temperature as in curves 1 and 2. This suggests that the
sudden drop in temperature is caused by the sudden evolution of vapors
(cooling) as occurs just prior to blowing (increased hissing noise heard
prior to explosions). All this suggests the use of a "hotter" acid system
to permit resin formation occuring at lower temperatures to permit more
strength development in the periphery while also reducing the stream pres-
sure in the panel.
Upon reviewing these data we became concerned that peanut hull waste com-
posites would not meet the 2400 psi modulus of rupture (flexure) require-
ment for structural qualification because of the lack of reinforcement with
the slenderness ratio of a fiber. Therefore, on the next full-scale test
it was decided to make a peanut hull panel conventionally with a urea resin
to determine if the reinforcement theory was correct. The results of this
evaluation are presented in Table 1-29-
The peanut hull furnish was prepared at Material Systems Corporation ( see
section on Type and Percent of Acid Absorbed by the Material - Peanut Hull
Waste and Table 1-3). The furnish was post dried to between 1.5 and 8%
moisture content and screened to all -20 material. The work at Washington
State University was started with 4 inch by 6 inch panels, Table 1-4, to
develop the previously discussed process parameters. The next step was to
fabricate 12 inch by 15 inch panels, Table 1-5, to further refine the pro-
cess.
For the purposes of this discussion a pressure relief cycle or a "bump
46
-------
*
Figure 1-11. Steam generation.
Figure 1-12. Panel removal
47
-------
Figure 1-13. Panel removal,
Figure 1-14. Panel cooling.
-------
Figure 1-15. Section cut for test.
Figure 1-16. Panel explosion.
49
-------
Figure 1-17. Debris from explosion.
-------
Figure 1-18. Segment of exploded panel,
-------
TABLE 1-29 CUT UP DIAGRAM AND PHYSICAL TEST RESULTS OF BOARD MADE WITH
UREA FORMALDEHYDE RESIN (PEANUT HULL FURNISH)
30
B
11"
B
2"
.1-.
MOR/MOE
Spec No. SG MOR(psi) MOB (psi x 10~6)
1 0.81 1100 0.270
2 0.81 1148 0.278
INTERNAL BOND
Spec No. SG PSI
1-A
1-B
2-A
2-B
0.81
0.81
0.81
0.81
115
122
122
122
52
-------
cycle" can be defined as rapid release, at a predetermined time, of the
pressure applied to the mat. The press plattens were slightly opened to
allow the vapor build up in the mat to be released. The pressure was then
reapplied and maintained until the press was completely closed.
"Decompression time" is a gradual reduction of the final pressure to 0 psi
over a predetermined time, after the press time selected was complete.
The closing pressure for each board was selected before pressing. This
pressure was reached within 0.5 minutes after pressing was begun. Once the
target pressure was obtained an additional 100 psi was added and the pump
turned off. As compression of the mat continued the pressure decreased.
This pressure drop was allowed to reach 100 psi lower than the target pres-
sure. At this point the pump was started and the pressure returned to the
100 psi above the target and the procedure repeated. Once the press was
closed (on stops) the pressure was reduced to about 200 psi in approxi-
mately 3 minutes and this pressure then held for any remaining time.
Board making data are presented in Tables 1-30 and 1-31 for the peanut hull
furnish. The tables are divided into runs depending on the board size made.
Table 1-30, covering the hammermilled peanut hulls, shows a trend from run
to run towards pressing conditions that are quite close to what might be
desired for making successful boards. Runs 5 and 7 indicate that conditions
are straddling the borderline between blown and intact boards. More inves-
tigation should be carried out to determine exactly what the optimum con-
dition would be. Table 1-31 presents the processing parameters for a
peanut hull waste panel fabricated with conventional processing techniques.
Each run will be briefly discussed below:
Run 1: Pressing at a relatively long press time (7 to 15 minutes), with
and without a bump cycle and without a decompression cycle caused blows
in all boards.
Run 2: A change from the 12 by 15 inch press to the 50 by 50 inch press for
this run created the same problem as noted in Run 1. The addition of a two
minute decompression cycle for Board 9 also resulted in a blow. A varia-
tion in the bump cycle was also introduced with Board 9.
Run 5: Board number 25, pressed with a nine minute press time, no bump
cycle and a two minute decompression cycle, blew. Reduction in the press
time to four and five minutes, increased pressure to insure closing of the
press, elimination of the bump cycle and the addition of a 1.25 minute
decompression cycle, yielded six intact boards and five blow boards. It
appears that this combination is in the "fringe" area of making successful
boards of this material and binder. It is quite encouraging, at this point,
to note the number of intact panels made. A further effort to concentrate
on optimizing the conditions should be carried out. o
Run 7: Board number 44, which was pressed for 3.5 minutes at 375 F, pro-
duced a severe blow during the decompression cycle. This blow occurred
when the pressure was reduced to 30 psi. Previous blows all occurred at a
pressure of 100 psi or less. Reducing the press temperature to 355 F and
keeping all other conditions the same appeared to be acceptable.
53
-------
TABLE 1-30 BOARD-MAKING CONDITIONS FOR PEANUT HULL FURNISH
U1
Board
No.
Run 1 -
I
2
3
4
5
6
Run 2 -
Weight
Formed
0.5 x 12
1900g
1900g
1900g
1900g
1800g
1800g
0.5 x 50
MC
x 15"
5.5
5.5
5.5
5.5
5.5
5.5
x 50"
(%) Temp
(°F)
Boards
375
375
375
390
390
390
Boards
Time
(min)
7
10
15
10
10
15
PSI Thick-
ness
in.
1000
1000
1000
1000
1000
1000
Density Boa
BlO
BlO
G^^Y
sto
BlO
min
Bio
Bio
Bio
5.5
7
8
9
55 Ibs 5.5 390
55 Ibs 5.5 390
55 Ibs 5.5 375
10
10
10
1000 0.55
1000
1000 0.45
Run 5 - 0.5 x 50 x 50" Boards
25
26
27
31
32
33
34
35
36
37
38
39
51
51
51
51
51
51
51
51
51
51
51
51
Ibs
Ibs
Ibs
Ibs
Ibs
Ibs
Ibs
Ibs
Ibs
Ibs
Ibs
Ibs
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
7
7
7
7
7
7
7
7
7
7
7
7
375
375
375
375
375
375
375
375
375
375
375
375
9
4
5
5
4.5
4
4
4
4
4
4
4
800
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
47
56
53
50
55
56
57
55
51
53
52
56
1.01
1.09
.92
1.07
1.03
0.98
0.93
0.89
0.93
1.01
0.94
1.01
0.92
Board Conditions
Blown, no bump
Blown, bump at 5 min, did not reach
stop
Blown, bump at 5 & 10 min, pop at 10
min
Blown, bump at 5 min, close at 7 min
Blown, bump at 5 min, close at 6 min
Blown, bump at 5 & 10 min, close at
Blown, bump at 3 & 7 min, blow at 7
min
Blown, bump at 2 & 5 min, blow at 5
min
Blown at 100 psi, bump at 3 min, hold
open for 2 min, back to pressure at
5 min
Blown, close at 7 min
Intact/blister, close at 3.5 min
Intact/blown, close at 3.5 min
Blown, close at 3.5 min
Blown, close at 3 min
Intact, close at 3 min
Intact/blister, close at 3 min.
Blown, close at 3 min
Intact/blister, close at 3 min.
Blown, close at 3 min
Intact/blister, close at 3 min
Blown, close at 3 min
(continued)
-------
Table 1-30 (Continued) BOARD-MAKING CONDITIONS FOR PEANUT HULL FURNISH
Ln
Ui
Board Weight MC(%) Temp Time
No. Formed (°F) (mj.n)
PSI
Run 7-
44
45
46
47
0.
25
25
25
25
25 x 50
.51bs
.51bs
.51bs
.51bs
X
5
5
5
5
50"
.7
.7
.7
.7
Boards
375
355
355
355
3
3
3
3
.5
.5
.5
.5
1200
1200
1200
1200
Thick-
ness
in.
0.24
0.33
0.28
0.32
Density Board Conditions
Blow at 300 psi, close at 1.5 min
0.84 Intact/blister, warp, close at 1.5 min
0.97 Intact/blister, warp, close at 1.75 min
0.82 Intact/warp, close at 2 min.
-------
TABLE 1-31 BOARD-MAKING CONDITIONS FOR PEANUT HULL FURNISH WITH UREA RESIN
Run 10 - 0.5 x 12 x 15 in. boards treated peanut hulls with 6% urea resin added
(Board pressed at a SG of 0.72)
Board Weight MC (%) Temp Time PSI Closing Board
Number Formed ( F) (min) Time Condition
(min)
28 1200g 7.3 325 5 500 1 Intact
29 1200g 7.3 325 5 500 1 Intact
30 1200g 7.3 325 5 500 1 Blown
56
-------
Table 1-32 presents the results of the tests for modulus of rupture,
modulus of elasticity, and internal bond according to the American Society
for Testing and Materials Standard D-1037, "Standard Methods of Evaluating
the Properties of Wood-Base Fiber and Particle Panel Materials". These
results indicate superior properties to that of a conventional process
(Table 1-29) with peanut hull waste. It is believed that the reinforcing
theory holds. Although the economics (discussed in the section Economics)
is acceptable, the strengths are less than the required 2400 psi modulus
of rupture. This makes the board acceptable for non-structural applications
but not qualifiable under structural codes.
Wood Waste—The Douglas Fir wood waste was treated with p-toluene sulfonic
acid at Washington State University in a commercial blender. The furnish
was dried to between 1.5% and 8% moisture content and screened to remove
all -20 mesh particles. The initial work, was done on 4 inch by 6 inch
panels. This data is presented in Tables 1-11, 1-12, and 1-13. The
processing information developed was used to produce 12 inch by 15 inch
panels. This data is presented in Tables 1-14 and 1-15.
The material was formed into mat and pressed, as described for peanut hull
waste in the preceeding section. Table 1-33 presents the board-making data
for Douglas Fir. This table summarizes the variations that were tried.
These included three binder treatments, combinations of binder treatments,
a range of moisture contents from 1.5 to 7.8%, press temperatures of 300 F
to 390 F, press times of six to fourteen minutes, bump and no-bump cycles
and 0 to 2.5 minutes of decompression time.
By the end of Run 9, the proper conditions for pressing were being approach-
ed. Only one of the final eight 0.5 by 50 by 50 inch boards blew. The
cause of this blow is not known as it was pressed under the same conditions
as the others.
The various runs using Douglas Fir will be discussed below:
Run 3: Preliminary work with 0.5 inch thick by 12 inch by 15 inch boards
was conducted to get a "feel" for how the material would react. The wire
screen mentioned under notes for Boards 19, 20, and 22 was placed on top
of the mat in an attempt to provide vent passages. However, other combi-
in the production caused blows.
Run 4: 0.5 inch thick by 50 inch by 50 inch boards were made with a
combination of binder treatments. Board 23, pressed at 300°F for seven
minutes produced an intact panel even though the press did not completely
close. Board 24, with the addition of one minute press time and a higher
pressure to close the press, blew.
Run 6: The moisture content of the furnish was increased from 3% to 7.8%
in an attempt to improve the plasticity of the material. Boards 40 to 41,
pressed at 325°F, did not reach the desired thickness and blew. Boards
42 and 43 were pressed at 340°F. The desired thickness was reached;
however, blows still occurred.
Run 8: In this run the board thickness was reduced from 0.5 inch thick to
0.25 inch thick. Three binder treatments, two moisture contents, two press
57
-------
TABLE 1-32 CUT-UP DIAGRAM AND PHYSICAL TEST RESULTS FOR BOARD 27 (PEANUT
HULL FURNISH)
Test
Panel
1- ^
i
4£
1
i"
4
1
t
14
\
8"
—
3-1
A
B
3-2
A
B
3-3
A
B
2-1
A
B
2-2
A
B
2-3
A
B
MOR/MOE
INTERNAL
x'
1-1
A
B
1-2
A
B
1-3
A
B
2" L 8" 12" 1 8" J 2"L_
.*_ - - 24" __
(continued)
58
-------
TABLE 1-32 (Continued) CUT-UP DIAGRAM AND PHYSICAL TEST RESULTS FOR BOARD 27
(PEANUT HULL FURNISH)
Spec
1-1
1-2
1-3
2-1
2-2
2-3
3-1
3-2
3-3
No.
MOR/MOE
SG MOR (psi) ^pli x 10
0.97 1427
0.96 1351
— —
0.89 597
0.01 1329
0.98 1410
0.93 918
1.06 1433
— — __
0.397
-
0.166
0.373
0.363
0.277
0.404
— —
Internal Bond
Spec No.
1-1A
1-1B
1-2A
1-2B
1-3A
1-3B
2-1A
2-1B
2-2A
2-2B
2-3A
2-3B
301A
3-1B
3-2A
3-2B
3-3A
3-3B
SG
0.92
1.00
0.99
1.00
0.99
1.00
0.88
0.87
1.04
1.03
0.96
1.01
0.89
0.93
1.04
1.05
1.05
1.08
PSI
59
105
158
155
156
152
44
37
169
197
163
177
44
53
179
204
223
238
59
-------
TABLE 1-33 BOARD-MAKING CONDITIONS FOR DOUGLAS FIR FURNISH
Binder
Number
Run 3 -
10
11
12
13
14
15
16
17
18
19
20
21
22
Run 4 -
23
24
Run 6 -
40
41
Weight
Formed
0.5 x 12 x
ISOOg
ISOOg
1600g
1600g
IGOOg
IGOOg
IGOOg
1600g
1600g
1600g
1600g
IGOOg
1600g
0.5 x 50 x
52 Ibs
52 Ibs
0.5 x 12 x
ISOOg
ISOOg
MC (%) Temp
15 in.
1.5
1.5
1.5
5.5
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
50 in.
3.0
3.0
15 in.
7.8
7.8
(F)
boards
390
390
390
390
300
300
300
300
300
300
300
300
300
boards
300
300
boards
325
325
Time
(min)
12
12
12
12
5
6
9
14
10
10
7
0.5
10
7
8
8
8
PSI Thick- Density Deo
ness pre;
Timi
750 - 0
1100 - - 0
1100 - - 0
1100 - - 0
1000 - - 0
1000 - - -
750 - - -
750 - - -
1000 - - 2.5
1000 - - 2
1000 - - 0
1100 - - 2.5
1000 - 2
1100 0.53 1.02 1
1200 0.58 0.95 1
1100 - - 0
1100
Comments
Blown, bump at 7 min in-
crease to 1500 psi, did
not close
Blown, bump at 7 min in-
crease to 1500 psi, did
not close
Blown, bump at 7 min, did
not close
Intact, bump at 7 min, close
at 5.5 min
No bump close at 4.5 min
Stuck to platten, close at
5.5 min.
Close at 7 min
Blow at 14 min 750 psi
Blown, close at 7.5 min
Blown, wire screen on top,
did not close
Blown, wire screen on top
Intact, no screen, close at
7 min
Blow at 300 psi, wire screen
on top, close at 8 min
Intact, did not close
Blown, close at 7 min
Blown, close & take out,
did not reach stops
Blown, bump at 5 min, did
not close
(continued)
-------
TABLE 1-33 (Continued) BOARD-MAKING CONDITIONS FOR DOUGLAS FIR FURNISH
Binder Weight
Number Formed
42 1800g
43 1800g
Run
48
49
50
51
52
53
54
Run
55
56
57
58
8 - 0.25
25.
25.
25.
25.
25.
25.
25.
MC (%) Temp
7.8 340
7.8 340
x 50 x 50 in,
5 Ibs
5 Ibs
5 Ibs
51bs
5 Ibs
51bs
5 Ibs
9 - 0.5 x 15 x
51
51
51
51
Ibs
Ibs
Ibs
Ibs
7.8
7.8
7.. 8
7.8
4.5
4.5
4.5
15 in.
4.4
4.4
4.4
4.4
. boards
325
325
325
310
310
310
310
boards
335
335
335
335
Time
(min)
10
10
6.5
6.5
6.5
6.5
6
6
6
9.5
10
10
9
PSI Thick-
ness
1100
1100
1100
1100
1100
1100
1100
1100
1100
1200
1200
1200
1200
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
28
27
25
26
28
30
32
39
55
54
55
Density
0.
1.
1.
1.
1.
0.
0.
0.
0.
1.
0.
98
02
08
04
01
93
89
93
98
01
97
Decom-
press:
Time
1
1
1.5
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.25
4.00
Blown, bump at 5 min,
blose at 9 min
Blown, bump at 5 min,
close 7 min
Blown, bump at 4 min,
close at 4.5 min
Blown, press break down,
repressed 2 days later -
bump at 5 min
Blow at 25 psi, bump at
5 min, close at 4 min
Blown, bump at 5 min,
close at 4.25 min
Intact, no bump, did not
close
Intact, no bump did not
close
Intact, no bump, did not
close
1.25 Blown, no bump, close at
8.5 min
1.25 Blown, bump at 5 min, close
at 8.5 min
1.25 Blown, bump at 8.5 min,
close at 8 min
4.00 Blown, no bump, press to
Back to stops and take out
pressure
for 30 sec. (continued)
-------
TABLE 1-33 (Continued) BOARD-MAKING CONDITIONS FOR DOUGLAS FIR FURNISH
en
Board
Number
59
60
61
62
63
64
65
66
Weight
Formed
51 Ibs
51 Ibs
49 Ibs
49 Ibs
49 Ibs
49 Ibs
49 Ibs
49 Ibs
MC (%)
4.4
4.4
4.4
4.4
4.4
4.4
4.4
4.4
Temp
335
335
335
335
335
335
335
335
Time
(min)
9
9
9
9
9
9
9
9
PSI
1200
1200
1200
1200
1200
1200
1200
1200
Thick-
ness
0.57
0.57
0.54
0.59
0.58
0.58
0.59
0.61
Density
0.97
0.96
0.98
0.89
0.91
0.91
0.91
0.87
Decom-
pression
Time
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.25
Comments
Intact, bump at 5.5 min,
close at 9 min
Intact, bump at 5.5 min,
close at 9 min.
Blown, bump at 5.5 min,
close at 9 min
Intact, bump at 5.5 min,
close at 9 min
Intact, bump at 5.5 min,
close at 9 min
Intact, bump at 5.5 min,
close at 9 min
Intact, bump at 5.5 min,
close at 9 min
Intact, bump at 5.5 min,
close at 9 min
-------
lines were used. Treatment 69-48-3 at 4.5% MS, 310°F, a six minute
press time without a bump cycle and a 1.25 minute decompression cycle
produced intact boards. However, the press did not completely close
on any of these.
Run 9: Twelve 0.5 inch thick by 50 inch by 50 inch boards were made in the
final run. The only variations between boards were the press time, bump
cycle, and the amount of furnish used. When the following conditions were
established a series of nine boards were made with only one blown panel
resulting. These conditions were: Binder Treatment 69-48-3, 51 and 40
pounds of furnish to form the board, 4.4% moisture content, 335°F press
temperature, nine minute press time, 1200 psi closing pressure, a bump
cycle of 5.5 minutes and 1.25 decompression period. The cause of the blow
is not known, but apparently the conditions were again in the borderline
category.
Table 1-34 presents the results of the tests for modulus of rupture,modulus
of elasticity, and internal bond according to the American Society for
Testing and Materials Standard D-1037, "Standard Methods of Evaluating the
Properties of Wood-Base Fiber and Particle Panel Materials'. The results
show that the material can qualify structurally but will require additional
full scale studies to develop product reliability- The economics show that
this process is equivalent to that of commercial board. Since there exists
neither a cost advantage nor a structural advantage, the product was not
pursued through qualification.
RECOMMENDED PROCESSING PROCEDURES
Peanut Hull Waste—Although further full scale development would be required
to productize this material, the following is presented as a starting point.
a. Screen the material to eliminate all particles - 20 mesh.
b. Use a commercial blender and mix with the hulls a solution of water
with 7% by weight by phosphoric acid (75% strength). Use 8 gallons
of solution for 8.82 pounds of hulls. Mix for five minutes.
c. Place the mix in a dryer, remove and save the solution for reuse. Dry
to 5.7% moisture content which will result in a 3.98% acid content.
d. Place in a pre-heated 355 F press.
e. Mold at 1200 psi for 3.5 to 4 minutes depending on thickness.
f. Close press within 2 minutes.
g. Use a 1.25 minute decompression time.
Wood Waste—Douglas Fir can be produced with the following procedure.
a. Screen the material to eliminate all particles -20 mesh.
b. Spray the wood particles with a solution of water containing 2.5% to
2.7% g"toluene sulfonic acid technical grade.
c. Place the mix in a dryer, remove and save the solution for re-use.
Dry to 4.4% moisture content which will result in a 1.05% acid absorp-
tion.
d. Place in a preheated 335 F press.
e. Mold at 1200 psi for 9 minutes.
f. Use a pressure relief cycle, "bump", at 5.5 minutes.
g. Use full 9 minutes to close press.
h. Use a 1.25 minute decompression time.
63
-------
TABLE 1-34 CUT-UP DIAGRAM AND PHYSICAL TEST RESULTS FOR BOARD 66 (DOUGLAS
FIR FURNISH)
66
•a * A f
™" 4c
Test
T^-» in. J-L *\
Panel
- 24" — ^
311 -
>,„,
j
4
8" 4f
I
14
J
J"
11
L.
3-1
A
B
3-2
A
B
3^3
A
B
2" * 8" --2" ^_ 8"
24"
(continued)
64
-------
TABLE 1-34 (Continued) CUT-UP DIAGRAM AND PHYSICAL TEST RESULTS FOR BOARD 66
(DOUGLAS FIR FURNISH)
Spec No.
1-1
1-2
1-3
2-1
2-2
2-3
\
3-1
3-2
3-3
MOR/MOE
SG
0
0
0
0
1
0
0
1
1
.84
.84
.83
.95
.00
.99
.96
.00
.00
Spec No
1-1A
1-1B
1-2A
1-2B
1-3 A
1-3B
2-1A
2- IB
2-2A
2-2B
2-3A
2-3B
3-1A
3-1B
3-2A
3-2B
3-3A
3-3B
MOR
1080
880
990
1830
2430
1970
1900
2130
2270
Internal
SG
0.82
0.87
0.84
0.83
0.82
0.82
0.97
0.97
1.02
1.06
0.98
1.00
0.96
0.98
1.02
0.99
1.00
1.02
MOE . -6
(psi x 10 )
0.257
0.222
0.245
0.455
0.607
0.564
0.532
0.582
0.590
Bond
PSI
28
42
32
32
33
36
84
88
99
107
86
87
90
96
95
94
106
105
65
-------
ECONOMICS
General--There are common factors to be used in both peanut hull and wood
waste particle board economic analysis. These factors will be discussed
first and then applied. This analysis is based on the references listed
in Appendix A.
Capital Investment—The cost of a particleboard plant in 1975 dollars is
approximately $3.2 million plus $135,000 per million square feet of annual
capacity. This cost is estimated to be accurate within 10% of the estimate
two thirds of the time.
Primitive plants of less than 100 million square foot capacity cost consid-
erably less than the estimate given above, probably from $1 million for a
3 million square foot plant to $3 to $4 million for a 10 million square
foot plant. The cost can vary a great deal depending on sophistication,
precise process, and whether new or used equipment is purchased.
The waste process will require five to eight times as much drying capacity
as conventional plants, an amount proportional to the increase in water
content of the wood to be dried. It will also require centrifuge to reduce
the water content.
The corrosive nature of the waste process may also require extensive use of
stainless steel or fiberglass in storage tanks, dryers, presses, blenders
and materials handling equipment. This requirement may cause a dramatic
increase in the capital cost of a plant.
Labor—The number of workers required per shift in a particleboard plant
can be calculated by starting with a base number of 21 and adding one
worker per 8,000,000 square feet of annual capacity. Small plants are
much more labor intensive than large ones.
Labor costs vary slightly with square feet and tonnage of board produced.
This report will assume constant labor costs per 1000 square feet of board
produced of $23.22.
Power—Power is directly related to tonnage of board produced and averages
250 KWH per 1000 sq. ft. of 3/4 inch 45# particleboard. Power is used
primarily in refining (flaking, chipping, blending, and drying) and in
pressing. Power cost assumed is $0.011 per KWH.
Heat —Heat is directly related to tonnage produced and averages 1,650,000
Btu (1,500 cu. ft. of natural gas) per 1000 sq. ft. of 3/4 inch, 45#
conventional particleboard. Heat is required for drying the incoming
wood and for heating the press. Btu required to raise temperature of water
in wood to 212 degrees in dryer:
66
-------
3000 Ib wood x .15 water x 153 Btu per Ib. = 68,850 Btu
Btu required to vaporize water in wood in dryer:
3000 Ib wood x .14 water vaporized x 970 Btu per Ib. = 407,400 Btu
Btu required to raise temperature of wood to 212 degrees in dryer:
3000 Ib wood x .5 specific heat x 153 Btu per Ib. = 229,500 Btu
Btu required to raise temperature of wood to 350 degrees in press:
3000 Ib wood x .5 specific heat x 250 Btu per Ib. = 375,000 Btu
Btu required to raise temperature of resin to 350 degrees in press:
3000 Ib wood x .06 resin x 250 Btu per Ib. - 45,000 Btu
Btu required to raise temperature of water in wood to 350 degrees in press:
3000 Ib wood x .07 water x 250 Btu per Ib. = 52,500 Btu
Btu required to vaporize water in wood in press:
3000 Ib wood x .05 water vaporized x 970 Btu per Ib. = 145,500 Btu
Total Btu calculated= 1,323,750
Total Btu required per industry data= 1,650,000
Process heat efficiency= 80%
Btu consumed in Drying 770,000 (47%)
Btu consumed in Pressing 880,000 (53%)
1,650,000
In the waste particle board process, the incoming wood at between 50 and
150 percent moisture is dried to 5 percent. It is then pressed at 350
degrees Farenheit where the moisture content is reduced to 2 percent.
The following is the-composition of 1000 square feet of 45 pound per cubic
foot density, 3/4 inch thick particleboard made by the waste process:
Material Raw Material Part by Finished Board,
Reqd, Ibs Weight Ibs
Wood 3,000 100 2,605
Binder 180 6 156
Water 1500 - 4500 2 52
4680 - 7680 108 2,813
Btu required to raise temperature of water to 212 degrees in dryer:
3000 Ib wood x 0.15 moisture x 153 Btu/lb = 68,850 Btu
3000 Ib wood x 0.50 moisture x 153 Btu/lb = 229,500 Btu
3000 Ib wood x 1.00 moisture x 153 Btu/lb = 459,000 Btu
3000 Ib wood x 1.50 moisture x 153 Btu/lb = 688,500 Btu
Btu required to vaporize water in wood in dryer:
3000 Ib wood x 0.10 moisture vaporized x 970 Btu/lb = 291,000 Btu
3000 Ib wood x 0.45 moisture vaporized x 970 Btu/lb = 1,309,500 Btu
3000 Ib wood x 0.95 moisture vaporized x 970 Btu/lb = 2,764,500 Btu
3000 Ib wood x 1.45 moisture vaporized x 970 Btu/lb = 4,219,500 Btu
Btu required to raise temperature of wood to 212 degrees in dryer:
3000 Ib wood x 0.50 specific heat x 153 Btu per Ib = 229,500 Btu
Btu required to raise temperature of wood to 350 degrees in press:
3000 Ib wood x 0.50 specific heat x 250 Btu per Ib = 375,000 Btu
Btu required to raise temperature of binder to 350 degrees in press:
67
-------
3000 Ib wood x 0.06 binder x 250 Btu per Ib = 45,000 Btu
Btu required to raise the temperature of water in wood to 550 degrees in
press:
3000 Ib wood x 0.05 x 250 Btu per Ib = 37,500 Btu
Btu required to vaporize water in wood in press:
3000 Ib wood x 0.03 specific heat x 970 Btu per Ib = 87,300 Btu
Total Btu Required in Waste Process
Moisture
15%
50%
100%
150%
Total Btu
Calculated
1,134,150
2,313,300
3,997,800
5,682,300
Btu Req'd at
80% Efficiency
1,420,000
2,890,000
5,000,000
7,100,000
Btu Used
in Drying
740,000 (52%)
2,210,000 (75%)
4,320,000 (86%)
6,420,000 (90%)
Btu Used
in Pressing
680,000 (48%)
680,000 (25%)
680,000 (14%)
680,000 (10%)
For purposes of this report, gas will be presumed to cost $0.50/100 cu. ft.
and the conventional process to utilize wood at 15 percent moisture (planer
shavings) while the waste process used wood at 100% moisture (planer shav-
ings soaked in acid bath and spun dry) or estimated 50% moisture (spray
technique).
The calculation of heat required does not include the heat needed to remove
humidity from the air coming into the dryers. Industry sources estimate
that a conventional particleboard plant will use twice as much heat on a
humid day as on a normal day. Since the conventional process uses about
47% of its Btu requirements to dry the incoming wood while a plant using
the waste process and wood at 100% moisture used 86% of its Btu require- ->
ment to dry the wood, humid days would increase the Btu requirements of a
waste plant to 285% of normal compared to 200% for a conventional plant.
Also not included in the Btu calculations are extra heat requirements to
cause the water to be released from the acid in the waste process, or the
effect of exotherm during pressing. These factors are believed to have a
small impact on total Btu requirements.
68
-------
Overhead—The following are overhead costs for a 60 million square
foot plant per year:
Salaries:
Plant Manager 33,000
Plant Supervisor 24,000
Technical Director 20,000
Bookkeeper 16,000
Clerk/Typist (2) 24,000
Shipping Clerk 16,000
Insurance 50,000
Property' Taxes 90,000
Office Expenses 80,000
220,000
353,000
Overhead per 1,000 square feet $5.88
Peanut Hull Waste'—The minimum density of acceptable particle board from
peanut hull waste is 55 pounds per cubic foot. This will be the density
used in the cost analysis. In this study the soak and dry process was
used to treat the peanut shells. It was possible to recover all but 5.8%
of the acid in one case and 6.8% in the other (Table 1-3). Further drying
at Washington State University reduced the acid absorption content to 4%.
It is reasonable to assume that this extra 2% could be recovered also with
a controlled process. Although not tried in this program, it is also
reasonable to assume that the spray process would work with peanut hulls as
well as with wood waste. Therefore, an acid absorption value of 4% is used
in the following cost analysis. In addition the spray process reduces the
moisture content to be removed from the wood resulting in lower fuel costs.
Assumptions:
3/4 inch board
15% scrap reprocessed
Conventional Board: 55# density - 8% resin
Urea Resin Cost - 10.5C/lb
Phenolic Resin Cost - 27C/lb.
Waste Process Board: Wood - 0.4£/lb
55# density - 4% phosphoric acid
Phosphoric Acid Cost - 18C/lb.
Peanut Hulls - .75
-------
Material Used (Pounds per 1000 square feet)
Conventional Waste Process
55# Urea 55# Phenolic
Peanut Hulls 3667
Acid 147
Wood
Binder
Wax
3667
293
37
3667
293
37
3997 3997 3814
Material Cost (Dollars per 1000 square feet)
Conventional Waste Process
55# Urea 55# Phenolic 55#
Wood 15.03 15.03 Peanut Hulls 27.50
Binder 30.77 79.11 Acid 26.46
Wax 1.85 1.85
47.65 95.99 53.96
If no additional resin or wax is needed to reprocess waste scrap and
normal amounts are required to reprocess conventional scrap, the
reprocessing costs are:
Material Cost of Scrap (Dollars per 1000 square feet)
Conventional Waste Process
55# Urea 551 Phenolic 55#
Wood Peanut Hulls
Binder 30.77 79.11 Acid
Wax 1.85 1.85
32.62 80.46 0
If 15% of the board produced is scrap, the materials cost of a finished
board is:
Material Cost of Finished Board (Dollars per 1000 square feet)
Conventional Waste Process
55# Urea 55# Phenolic 55#
Primary 40.50 81.59 45.86
Run 85%
Scrap Re- 4.89 12.14
Run 15%
45.39 93.73 45.86
70
-------
The total manufacturing costs in a 60,000,000 square foot particleboard
plant are:
Item
Variable Costs
Materials
Power
Fuel
Labor
Maint & Supplies
Conventional Waste Process 55#
55# Urea 55# Phenolic Soak Test Spray Test
45.39
3.36
.92
23.22
2.44
75.33
93.73
3.36
.92
23.22
2.44
123.67
45.86
3.36
2.77
23.22
2.44
77.65
45.86
3.36
1.60
23.22
2.44
76.48
Fixed Costs
Overhead
Depreciation
Total Costs
5.88
18.83
24.71
100.04
5.88
18.83
24.71
148.38
102.36
5.88
18.83
24.71
101.19
From this analysis it is apparent that no economic advantage over conven-
tional wood particle board exists with peanut hulls. However, in the event
that large supplies of the waste exists, it could be an advantageous method
for waste utilization.
Wood Waste—The minimum density of acceptable particle board from wood waste
is 55 pounds per cubic foot also. The spray technique for treating the wood
waste was demonstrated to be acceptable. The p-toluene sulfonic acid is
used at an absorption content of 1.05% of the wood weight. All other acids
can be removed.
Assumptions:
3/4 inch board
15% scrap reprocessed
Conventional Board: Wood - 0.4C/lb
55# density - 8% resin
Urea Resin Cost - 10.5*/H>
Pehnolic Resin Cost - 27£/lb
Waste Process Board: 55# density - 1.05% p-Toluene Sulfonic Acid
p-Toluene Sulfonic Acid Cost - 23£/lb
71
-------
Material Used (Pounds per 1000 square feet)
Conventional Waste Process
55# Urea 55# Phenolic 55#
Wood 3667 3667 Wood Waste 3667
Binder 293 293 Acid 39
Wax 37 37
3997 3997 3706
Material Cost (Dollars per 1000 square feet)
Conventional Waste Process
55# Urea 55# Phenolic 55#
Wood 15.03 15.03 Wood Waste 15.03
Binder 30.77 79.11 Acid 8.97
Wax 1.85 1.85
47.65 95.99 24.00
If no additional resin or wax is needed to reprocess waste scrap and
normal amounts are required to reprocess conventional scrape, the
reprocessing costs are:
Material Cost of Scrap (Dollars per 1000 square feet)
Conventional Waste Process
55# Urea 55# Phenolic 55#
Wood - - Wood Waste
Binder 30.77 79.11 Acid
Wax 1.85 1.85
32.62 80.46 0
If 15% of the board produced is scrap, the materials cost of a finished
board is:
Material Cost of Finished Board (Dollars per 1000 square feet)
Conventional Waste Process
55# Urea 55# Phenolic 55#
Primary 40.50 81.59 20.40
Run 85%
Scrap Re- 4.89 12.14
Run 15%
45.39 93.7320.40
72
-------
The total manufacturing costs in a 60,000,000 square foot particleboard
plant are:
Total Cost (Dollars per 1000 square feet)
Item Conventional Wood Process
55# Urea 55# Phenolic 55#
Variable Costs
Materials 45.39 93.73 20.40
Power 3.36 3.36 3.36
Fuel .92 .92 1-60m
Labor 23.22 23.22 41.80
Maint & Supplies 2.44 2.44 2.44
75.33 123.67 69.60
Fixed Costs
Overhead 5.88 5.88 5.88
Depreciation 18.83 18.83 18.83
24.71 24.71 24.71
Total Cost 100.04 148.38 94.31
The time to process the panel is 9 minutes rather than 5 minutes used
conventionally. This labor cost could be reduced by using a press with more
openings but this would increase capital cost and depreciation cost.
This product has limited economic advantages and is limited to a minimum
density board of 55 pounds per cubic foot in a market where the predominate
sales are with 45 pounds per cubic foot. The effort on this program was
not sufficient to develop full processing parameters. Although the limita-
tions are known, not all of the potentials have yet been established.
PROCESSING WITH DOUGLAS FIR FINES
Past work with peanut hulls and wood particles indicated that the fines
from either generated more resin than larger particles; this in turn caused
processing problems because of excessive shrinkage and/or blowing or
blistering from excessive steam evolution. It was proposed to study the
fines as a source of resin but in the presence of other materials to act
as reinforcement and to permit escape of steam.
First efforts were to examine the fines alone to determine the problem
areas. Douglas Fir, hammermilled material, was screened and all the -20
mesh material used. Three formulations were made, each soaking 1000 g of
fines in a p-toluene sulfonic acid solution made from 150 g p-toluene
sulfonic acid and 2500 g of water (I) , 1125 g p-toluene sulfonic acid in
2500 g of water (II) and 75 g p-toluene sulfonic acid and 2500 g of water.
After soaking for 30 minutes the material was spun in a centrifuge to re-
move most of the acid solution. The wet material was dried until a
73
-------
moisture content of about 7% was achieved. The data are shown in Table
1-38.
The data in Table 1-38 indicate some outstanding strength retention after
24 hour water soak with minimal water pickup. Again this is achieved at
densities higher than commercial high density boards and at a cost for
acid which makes cost greater than present commercial material. However
this may be mitigated under certain conditions and should not be excluded
from further considerations where price is not the critical factor.
Again the data showed that the amount of acid picked up by the wood is
crucial to determining strength dry and wet. With an acid pickup of
2.8% the final product soaked up so much water as to render it unhandle-
able or testable. An acid pickup of 4.2% gave excellent results while the
5.7% acid pickup gave outstanding properties.
The very poor results with 1/8" chopped glass added was due to several
factors with poor wet out of glass and wrong amount of glass being the most
critical factors visible.
Due to time remaining on the program and the need of funds for more
promising products, further work with the particle board products was halted.
RECOMMENDATIONS AND CONCLUSIONS
Peanut Hull Waste-—The resulting product is limited to a minimum weight
of 55 pounds per cubic foot and without the benefit of some fibrous rein-
forcement will not be able to achieve a modulus of rupture of 2400 psi.
However, values of 1400 psi to 2100 psi (see Tables 1-4 and 1-32) are
respectable. The economics of the product are marginal based on the present
price of peanut hulls. This price would change dramatically if there were
large surplus quantities of hulls which required disposal. In the United
States peanut hulls have many uses and are a marketable waste. However, in
many of the developing nations peanuts are major crops with oil as the
primary product. It is these nations which could make use of this product.
Assuming that interest exists in the product, there exists at least one
year of process development on full-scale equipment. This program
demonstrated a process that can make 50 inch by 50 inch panels although
not consistently. It is believed that the acid content and moisture content
have been established. Further studies are required on
a) Particle size and distribution
b) Pressures
c) Pressure relief and decompression cycles, and
d) Fiberous reinforcements
It is recommended that the Environmental Protection Agency make this process
available to the Department of State for exploitation abroad.
Wood Waste—This product is limited to a minimum weight of 55 pounds per
cubic foot also. Modulus of rupture values in excess of 2400 psi were
74
-------
TABLE 1-35 CHIP BOARD PRODUCT USING DOUGLAS FIR FINES WITH P-TOLUENE SULFONIC ACID
Batch
I
I
I
II
II
II
III
IIIA
IV
% Conc'n
Acid
Solution
6.0
6.0
6.0
4.5
4.5
4.5
3.0
3.0
6.0
% Acid
Pickup
by Wood
5.7
5.7
5.7
4.2
4.2
4.2
2.8
2.8
5.7
Moisture
Content
%
7.3
7.3
7.3
7.8
7.8
7.8
7.6
7.6
7.6
Time
Mins
5
10
5+5
5+5
15
5
5
5
5
Temp
°F
300
300
300
300
300
325
300
300
300
Pressure
psi
750
750
500
500
500
500
500
500
750
Density
1.05
1.30
1.18
1.16
1.26
1.18
1.08
1.24
1.18
Flexure MOR
psi
849
3486
2874
3296
3176
3171
2044
3204
2402
-------
achieved (Tables 1-12 and 1-34) although not consistently. The product has
some economic advantage. This advantage could be increased by judicious
process development work. This program demonstrated a process that can make
50 inch by 50 inch board somewhat consistently (see Run 9 - Table 1-33).
However, additional process development is required on full scale equipment
to define
a) Particle size and distribution
b) Pressures, and
c) Pressure relief and decompression cycles.
It is, therefore, mandatory that an existing particle board manufacturer
become interested in the process in order to further this development.
76
-------
SECTION 3
WASTE PRODUCTS USING AN INORGANIC MATRIX
INTRODUCTION
The Phase II study developed two classes on inorganic materials. One was
an inorganic matrix foam using rice hulls. The other was a dense inorganic
matrix reinforced with wood waste and/or rice hulls as a lumber substitute.
The rice hull foam could be produced from a variety of formulations. Within
the denser range products suitable for structural walls, floors, ceilings, and
combination floor-ceilings could be produced with a possible two or three hour
fire rating. On the opposite end of the density scale, a rice hull foam
could be produced which could be used as a core for a fire door.
The purpose of Phase III was to scale up the Phase II data and to produce
full scale products, test and qualify them. The products evaluated are
discussed below.
WALL SYSTEM
Material Selection— There were a variety of rice hull foam formulations
evaluated but the one selected for qualification is presented in Table 2-1.
The MSC additive improves the water resistance of the material. Once the
material has cured, the moisture content stabilizes at 10%. Then the weight
and cost distribution of the foam is as shown in Table 2-2.
TABLE 2-1 RICE HULL FOAM FORMULATION
Constituent Amount, Ibs
Water 100
Rice Hulls 57
Untreated Glass Fibers 3.4
MSC Additive 5
Plaster 343
Bussan 30 0.6
77
-------
TABLE 2-2 CURED WEIGHT AND COST DISTRIBUTION
Constituent
Water
Rice Hulls
Untreated Glass Fiber
MSC Additive
Plaster
Bussan 30
Total
Cured Weight
Ib
46
57
r 3.4
5
343
0.6
Cost/lb
_
$0.62 *
.40
.432 *
.0252
2.75
Total
Cost
_
$3.534
1.360
2.160
8.644
1.6500
455
.038
17.348
Price delivered to factory.
78
-------
This formulation was selected for its properties and resistance to water.
The effects of water on the strength retention of rice hull foam was
studied by constructing a panel and cutting it into 144 compression blocks,
1" x 1" x 2" in size. Thirty-six of the blocks were uncoated. The remain-
ing blocks had their exterior surfaces coated with one of the following
three materials.
a. Swimming Pool Paint produced by Zynolite Products Corporation
b. Lytron 621 produced by Monsanto
c. W-102 produced by Wetco Chemicals
Nine each of the four classes of specimens were immediately tested as
controls. Eighteen of each class were placed into separate jars of water.
These specimens were soaked in the same water throughout the test period.
The remaining nine specimens of each class were placed in separate jars
of water in which the water was changed daily. The purposes of the tests
were to determine the following:
a. The effects of one day, one week, and one month of complete submersion
in water on strength retention.
b. The effect of standing water versus transient water on strength
retention.
c. The effects of surface coating on strength retention.
d. The ability of the material to regain its strength after drying out.
Samples of kiln dried Douglas Fir, a standard building material, were
subjected to the same tests as a comparison.
The tabulated data for these tests are presented in Appendix B. Figure
2-1 presents a graphical summary of these results. A study of these
curves will establish the following conclusions:
a. The uncoated sample performed as well or better than the coated ones.
b. The major reduction in strength when tested wet occurs the first day.
After that, over a month's period of submersion, there is very little
change. In all cases there appeared to be even a slight regain in wet
strength.
c. W-102 coating appears to have an adverse effect.
d. Up to one week submersion there appears to be a 100% regain in strength
on all but W-102 coated samples, when the material dries out. After
one month submersion strength retention upon drying out appears to be
between 82% and 97%0
Figure 2-2 compares the uncoated samples strength retention against the
strength retention of Douglas Fir, a common building material. It is
amazing that the wet strength results on a percentage basis are nearly
identical. Unfortunately the effect of one day submersion on wood was not
determined 0
The average wet compression load after 30 days submersion is slightly over
300 psi. The area of a linear foot of wall as now designed is 15.5 square
inches. Thus the wall wet after 30 days submersion will support up to 4650
pounds per linear foot. A typical maximum wall design load would range from
800 pounds to 1200 pounds per linear foot. Therefore, this wall would pro-
vide the required Factor of Safety of 3 with some additional margin.
79
-------
0 0 Tested wet after soak in static water.
e $ Tested wet after soak in water change daily.
.j. j. Tested dry after soak in static water.
Uncoated
600 J
f
400 -
200 -
600 ,
400 -
200 ..
H
g 600 -
1 400 -
c
.3 200 -
w
CO
a>
o
u
IL 4 -
V- "Jj___
{y* — ' — ~~ "" "^
iJiij'i (ii»i l'J' "
lO 20 30
Coated with Swimming Pool Paint
1
«T " • — •£
?
10 2*0 ' ' ' ' 3*0
Coated with Lytron 621
___^ — -L _
W— = — 2 -S
'
io 2*0 3'o
600
400
200
Coated with W-102
1—i—•—t—1—I—-
10 20
Days in Water
i—r
Figure 2-1. Effects of water on strength retention.
80
-------
§
C
a
&
4J
W
100 .
90 .
80 "
70 '
60
50
40 '
30
20 -
10
Uncoated Rice Hull/Inorganic
Tested Dry
\\
1 \
I \
I >
1
Kiln Dry Douglas Fir
Tested Wet
O--
L- Uncoated Rice Hull/Inorganic
Tested Wet
~\
10
• — i
20
30
Days in Water
Figure 2-2. Percent strength comparison of rice hull/inorganic and wood.
81
-------
The mechanical properties of the foam were evaluated by Testing Engineers
of San Diego. The mean results are summarized in Table 2-3 and the detail
data are presented in Appendix C.
The first sample population for compression testing was too small for a
foam material and had to be retested with a larger population. The test
scatter is a result of the size of the specimen which is small compared to
the variation in porosity distribution from the rice hulls. The tensile
value appears low and in all reality would be much higher if evaluated in
a four foot wide specimen rather than one with a one inch cross section.
However, 54 psi is, adequate when the maximum wall tensile load probable for
a house is 115 Ibs per linear foot. The wall cross section of 15.5 square
inches would result in a tensile stress of 7.4 psi providing a safety factor
of 3 with considerable to spare.
Design Selection—The initial work in Phase II on full scale walls utilized
a molded shape bonded at the center line with an inorganic adhesive. This
is shown in Figures 2-3 and 2-4. These panels required considerable materials
and expensive fiberglass. The panels always failed in the center borderline
as could be expected since this is the point of maximum shear. Therefore,
an attempt was made in this phase to produce a monolithic wall having a
thickness of 3 1/2 inches.
TABLE 2-3 MECHANICAL PROPERTIES OF RICE HULL FOAM
Property psi
Compression Strength 952
Tensile Strength 54
Modulus of Rupture 165
Shear Strength 1,361
82
-------
Figure 2-3. Cross section of Phase II wall,
-------
^^^^
Figure 2-4. Full-scale Phase II wall,
84
-------
The mold was made in such a way that inserts were used to form the ribs and
void space and these were removed by slowly forcing out the inserts. Despite
the generous use of wax, as a parting agent, it was not possible to remove
the inserts without damage to the final structure. The design was then
changed to produce elements as shown in Figure 2-5 and 2-6. The internal
dimension of 3 1/2 inches was chosen for compatibility with conventional
construction. These panels could be extruded or cast continously by machine.
The length selected for this study was 8 feet. The elements for this study
were cast easily on the tool shown in Figure 2-7. The elements were then
bonded together to form the wall as shown in Figure 2-8.
Compression Strength of Wall— The first compression test conducted to verify
this design was on a two foot wide by four foot long specimen shown in Figure
2-9 and 2-10. This resulted in a failure load of 6000 pounds per linear foot.
This was then followed by three full scale qualification tests on four foot
wide by eight foot long panels. A diagram of the compression test set up is
shown in Figure 2-11 and Figure 2-12 shows a panel ready for testing. A
representative of Testing Engineers-San Diego witnessed the fabrication and
bonding of all panel components.
The sample panels were tested at Material Systems Corporation's plant,
Escondido, California. The compressive load was applied by means of a
hydraulic system recently calibrated by Testing Engineers-San Diego. The
test set-up was as described in ASTM E-72-74 (a) and illustrated in Figures
2-11 and,2-12.
85
-------
3/8"
00
en
3 1/2"
1/2"
~
S./R"
2' -O1
Figure 2-5. Cross section of wall material.
-------
Figure 2-6. Rice hull foam wall elements.
Figure 2-7. Tooling for scale-up studies.
87
-------
oo
OD
Figure 2-8. Assembly of wall material into a panel.
-------
Figure 2-9. Cross section of specimen.
-------
Figure 2-10. Compression failure.
-------
Hydraulic Ram & Load Point
Test Fixture
Loading Beam
Hydraulic
Reservoir
Axial Compression
Test Panel
Deflectometer
Mirror, Scale and
Wire
Lateral Guide
Dial Indicators
Figure 2-11. Material Systems Corp. panel axial
compression test set-up.
-------
Figure 2-12. Full-scale compression test,
.
-------
Lateral buckling of the panel was measured at mid-height by 0.01 inch taut
wire deflectometers on each panel edge. Axial compression of the specimen
under load was measured by means of 0.001 inch steel rod compressometers on
each of the four corners of the panel. The panel was pre-loaded to approx-
imately 400 pounds prior to setting deflection gauges. The loading was then
applied in increments of approximately 600 pounds in test SD31-1226 and 900
pounds in test SD31-2043 until failure occured. After each loading increment
the load was returned to zero and set measurements were made.
The details of these tests are presented in Appendix C. Figure 2-13 presents
the relationship of load to deflection and permanent set. Test SD31-1226 was
the first full scale panel produced and tested. Tests SD31-2043 were panels
produced in a simulated production run and were selected at random. It is
believed that these later panels represented a better quality product and this
is the reason for a more stiffer product and one with less permanent set.
In either case, the wall is considerably stiffer than other qualified systems
and exhibits less permanent set. The qualifiable design allowable is the
ultimate divided by a factor of safety of 3 and the width of the panel.
Panel SD31-1226: 31,800/3 x 4 = 2,650 Ib/ft
Panel #1 (SD31-2043): 30,000/3 x 4 = 2,500 Ib/ft
Panel tt2 (SD31-2043): 28,000/3 x 4 = 2,333 Ib/ft
The test curve indicates that the panel was within its elastic limit. There-
fore, the certified allowable will probably be 2,333 pounds per foot which is
approximately twice that of a stud wall.
Racking Shear Strength of Walls—These tests are also required for structural
wall qualifications. A 48 inch by 96 inch panel was fabricated per Figure
2-14 for a racking shear test. Because of the low tensile strength of the
material it was decided to incorporate a 0.040 inch thick fiberglass strip
at the lower bottom corner to assist in transfering tension at that point.
The upper and lower sill plates are 2" x 4" lumber bonded with epoxy to the
panel.
The panel was tested in the Material Systems Corporation's fixture shown in
Figure 2-15. The results of this test are shown in Figure 2-16. The defle-
ction under load shows that this panel is considerably stiffer than an ICBO
approved panel now in production. Also the failure load is approximately
four times higher than that of the approved panel. The allowable racking
shear load is the lower of the load that creates 1/8 inch lateral deflection
or the failure load divided by a factor of safety of 3. In this case both
values are the same, 450 pounds per foot for a four foot wide panel. The
failure of this panel was one of pure theoretical shear in which a shear
crack formed diagonally across the panel. These test results were sufficient-
ly high to encourage a repeat test without the strips.
93
-------
40,000 J
30,000
8
20,000 .
§
H
W
CO
8
10,000
—1
0.05
Failure 28,000 Ibs
Failure 30,000 Ibs
800 Ibs
DEFLECTION-INCHES
PERMANENT SET-INCHES
Figure 2-13. Compression load vs deflection and permanent set
for a 4 ft X 8 ft rice hull foam fire wall.
94
-------
k L'
ITTi
TT
T
96"
FIBERGLASS STRIP
\
r- 1
/ /
1 \
1
48"
Figure 2-14. Wall specimen.
95
-------
HYDRAULIC RAM AND LOAD POINT
HYDRAULIC JACK
DIAL INDICATOR
SPECIMEN
DIAL INDICATOR
DIAL INDICATOR
Figure 2-15. Material Systems Corp.
panel racking shear test
-------
4000
I
i
3000
2000
1000 .
with reinforcing
without reinforcing
strips
0.1
0.2
0.3
DEFLECTION-INCHES
Figure 2-16. Racking shear test on panel without reinforcement.
97
-------
A panel was again fabricated per Figure 2-14 without reinforcing strips and
tested in racking shear. The results are presented in Figure 2-16. Although
this panel provided results slightly lower (failure 48,000 pounds compared
to 54,000 pounds and 1/8 inch deflection at 14,000 pounds rather than 18,000
pounds) the design values are still higher than that generally found in a 48
inch wide shear panel. Failure started with a slight crack in the lower
tension corner which was reinforced with the fiberglass strip in the previous
tests.
Based on the above results three more panels were fabricated under Testing
Engineers'-San Diego observation per Figure 2-14 without reinforcement for
testing under racking shear by them in the Material Systems Corporation's
facility.
The racking shear loading was applied by means of a hydraulic system recently
calibrated by Testing Engineers-San Diego. The panels were mounted in a
test frame and deflection measured as outlined in ASTM E72-74.
Pre-loads of approximately 400 pounds were applied to the racking shear panels
and dial gauges reset prior to testing. The loading was then applied to the
racking shear test panels in 200 pound increments until failure occurred.
After each increment of loading the load was returned to zero and set
measurements were made.
The details of these tests are presented in Appendix C. Figure 2-17 presents
the relationship of side load to deflection and permanent set. The quali-
fiable allowables are established on the lower of the values derived from
ultimate load or at 1/8 inch deflection. These are discussed below.
(a) Ultimate Load Basis ,
The ultimate load has to be divided by a factor of safety of 3 and by
the width of the panel'to achieve a design allowable load per foot.
Panel #1: 3000 pounds/3 x 4 = 250 pounds/foot
Panel #2: 2800 pounds/3 x 4 = 233 pounds/foot
Panel #3: 3000 pounds/3 x 4 = 250 pounds/foot
(b) Load at 1/8 inch Deflection
The load at 1/8 inch deflection can also be used as the allowable
provided it does not exceed that of the above.
Panel #1: 1,150 pounds/4 = 288 pounds/foot
Panel #2: 1,300 pounds/4 = 325 pounds/foot
Panel #3: 1,350 pounds/4 = 338 pounds/foot
The calibrated equipment was limited to a force of 3000 pounds which was not
sufficient to fail two of the three qualification panels. However, the fail-
ure of one panel at 2,800 pounds would off set any higher loads developed by
the two other panels.
98
-------
3000 J
No 1, No 3
to
•a
g
•o
•H
W
H
(d
+J
O
EH
2000
1000
(Failure 3000 Lbs)
.——•
(Failure 2800 Ibs)
(Failure 3000 Ibs)
0.05 0.10 0.15 0.20 0.25
Deflection - Inches
Permanent Set - Inches
Figure 2-17. Racking shear load vs deflection and permanent set.
-------
The load allowable that will be certified by ICBO will probably be 233
pounds per foot. The load can be raised by repeat testing of three more
panels assuming that these three all fail in a close and similar pattern.
However, the 233 pounds per foot is a very acceptable load when compared
to standard constructed four foot wide walls. It is, therefore, not consid-
ered to be worth the expense to pursue this further.
The allowable based on deflection for panel #3 was close to that established
by earlier tests. However, the other two values were 25 to 60 pounds per
foot lower.
The allowable load based on ultimate failure at 2,800 pounds (233 pounds/
foot) is lower than that determined by earlier tests. However, the earlier
tests developed a crack at 3,000 pounds but still supported loads up to
4,800 pounds. This would indicate that the safest criteria would be to accept
the 2,800 pounds as the limit.
Impact Load Test of Walls—These panels were fabricated per Figure 2-14 under
the supervision of representatives from Testing Engineers-San Diego. Three
panels were tested utilizing the testing techniques described in ASTM E-72
with the test specimen mounted horizontally. Three specimens rather than
six (as indicated in ASTM E-72) were tested since construction of these panels
is such that interior and exterior faces are the same.
Each test was conducted as follows: Initial drop of the 60 pound sand bag
was made from a height of six (6) inches above the geometric center of upper
panel face. Subsequent drops were made from increasing increments of six
(6) inches until the panel being tested exhibited visual evidence of failure.
Deflection readings were made after each one-half foot increment of drop.
Set readings were made after each one-foot increment of drop.
The results were as follows.
Panel No. 1: Maximum deflection was 0.65 inch and maximum set 1.85 inches.
Failure was in the form of complete fracture through the panel section at a
drop height of .2;5 feet.
Panel No. 2: Maximum deflection was 0.60 inch and maximum set 1.32 inches.
Complete fracture through the panel section (at about mid-span) occurred
at a drop height of 2.0 feet.
Panel No. 3: Maximum deflection 0.58 inch and maximum set 1.27 inches.
Failure occured at a drop height of 2.0 feet.
Fire Rating—The resistance of the wall to fire was first evaluated at
Material Systems Corporation with a small fire chamber. This chamber is
capable of burning a 2 foot by 3 foot surface area of material. Tu save
money, a section of the first floor panel tested was cut into a fire test
specimen (2 Feet wide and 4 feet high).
This was designated panel #1. The surface had a crack in it from the beam
test but was considered to be adeauate for a fire test.provided the uncracked
surface was exposed to the fire.
100
-------
The results from this test defined as #1 are presented in Figure 2-18.
Unfortunately half way through the test it was determined that the cracked
surface was exposed to the fire. This crack was observed through the furnace
opening to separate at about one hour into the test causing what we believe
the backside temperature to reach in excess of the allowable at one hour
and fifty-five minutes.
A two foot wide by eight foot long section (see Figure 2-19) was taken from
the undamaged part of a racking shear panel and designated panel #2. The
results from this test designated as #2 are also shown in Figure 2-18. There
were furnace difficulties which prevented the achievement of a full ASTM
E-119 test. However, the results are conclusive enough to continue with a
full scale two hour fire test. Specimen #2 was tested in compression after
the fire test and failed at 3,000 pounds per linear foot. This is more than
the qualifiable design load.
The results of this test indicated sufficient capability to proceed with a
full-scale fire test.
Elements were produced for two 8 foot high by 12 foot wide walls and trans-
ported to the University of California Richmond Facility.* These elements
were assembled into panels, and mounted into test frames, Figure 2-20. The
endurance panel was instrumented with nine thermocouples as shown in Figure
2-21. It was placed in front of a furnace and loaded in'compression to 1000
pounds per linear foot by hydraulic jacks shown in Figure 2-22. After 1 hour
and fifty minutes of fire endurance testing per ASTM E-119, the panel lost
ability to support load and the upper right hand thermocouple exceeded 250 F
above ambient. This high temperature was caused by a crack in the back panel
caused by the buckling in the upper right corner shown in Figure 2-22. The
unexposed surface was otherwise intact. This premature buckling was caused
by the inadequacies of the bond of the stiffeners to the back surface. This
is shown clearly in Figure 2-23.
For further information see Fire Test Report: Fire Test of Material
Systems Corporation's One Hour Load Bearing Wall Assembly. University
of California, Berkeley, 1977.
101
-------
2000
I
M
&
1000
0 Test #1 Furnace Temperature
Test #1 Back Side Temperature
Test #2 Furnace Temperature
X Test #2 Back Side Temperature
Max. Acceptable Back Side Temperature
1 HR
2 HR
Figure 2-18. E-119 fire wall test.
102
-------
Figure 2-19. Small section fire test.
103
-------
Figure 2-20. Mounting walls into test frame.
Figure 2-21. Instrumentation,
104
-------
Figure 2-22. Buckling during endurance test,
105
-------
Figure 2-23. Stiffener bond failure to unexposed surface.
Figure 2-24. Exposed surface upon removal from furnace,
106
-------
Since the endurance panel did not achieve two hours it was decided to attempt
to qualify the wall for one hour by conducting a one hour rating hose stream
test on the remaining panel. This test consists of running a fire endurance
test for one half hour, spraying a hose stream over the surface for one minute
and then increasing the load to twice that of initial loading.
Figure 2-24 shows the exposed surface of the panel upon removal from the
furnace. Figure 2-25 shows the application of the hose stream. Figure 2-26
shows the wall appearance after the hose stream test and before increasing
the load by a factor of two. Upon the increase of load, buckling failures,
as shown in Figure 2-22, occurred at 1900 pounds per foot or 100 pounds less
than required for qualification. This resulted from similar bond failures
as that shown in Figure 2-23. From these tests, it was determined that a
solution would be to increase the bond area of the stiffener. This is shown
in Figure 2-27 in which these panels are being readied for a second fire
test. Also a surface coat of magnesium oxy-chloride was applied.
Three panels, Figure 2-28, 8 feet by 12 feet were made for the second series
of tests. The first panel was tested for a two hour rating hose stream test.
This test consists of submitting the panel to the endurance test under load
for one hour, applying a hose stream to the exposed surface, and then doubling
the load. The previous panels tested were not coated which permitted the rice
hull foam to vent any steam generated internally. This panel was coated and
had no vents in the wall. When heated, the steam generated internally was
sufficient to blow a section out of the panel. The test was continued, the
hose stream penetrated the wall opposite where the segment was blown out.
The second panel was vented with 1/4 inch holes in the lower and upper plates
and was tested as an endurance panel. > The wall supported load for two hours
and the highest backside temperature was 197 F which is considerably below
the 320 F permissible.
The third panel was a repeat of the hose stream test for a two hour certifi-
cation.
This test consists of submitting the wall to the fire under load for one
hour, then removing it, submitting it to a hose stream, permitting it to dry,
and then doubling the load. Upon testing, the hose stream penetrated the
wall although it did sustain the load.
Review of the test specimen showed that the material did not receive suffici-
ent heat to calcify the back surface. It is believed that stress concentrat-
ions were caused by the magnesium oxide chloride surface coat interaction
with the plaster of the rice hull foam when it expands the surface. These
stresses would be particulary high near the intersection of the surface panel
and the rib. When impacted by the normal load from the fire hose, these
combined stresses could be sufficient to cause localized failure.
A panel was fabricated using the modified surface coating application tech-
nique. This panel was subjected to and successfully passed the hose stream
test for one hour fire wall qualification. A two-hour qualification is
107
-------
i
Figure 2-25. Hose stream test.
Figure 2-26. After hose stream test.
108
-------
M
Figure 2-27. Panels with new rib.
Figure 2-28. Three panels ready for shipment.
109
-------
possible but will require further full-scale testing at considerable expense.
Since one hour is all that is required in 90% of construction, a two hour
hose stream test study at this time does not appear warranted. Therefore,
the wall will withstand a fire and support load for two hours but will permit
penetration by a stream of water from a hose after one hour. Complete details
on fir tests are presented in Appendix D.
Wall Thermal Characteristics—The test equipment used was the calorimeter
shown in Figure 2-29, a Leeds-Northrup Speed-C—Max W recorder with 24 therm-
ocouple probe readout, Shimadzu Model R-101 with 2 junction differential
iron constant thermocouple, Adjust-A-Volt 230 volt variable auto trans-
former connected to 4 black body heaters, and iron-constantan thermocouples.
Tests were run with a uniform input to the heating elements and this was
measured and monitored by a thermocouple placed on the hot face of the test
specimen while the flow, of differential, was measured by an intermediate
thermocouple in the center and another on the back side. The test was
allowed to run for an 8 hour period where temperatures were marked at
hourly intervals on the hour. At the end of this period a hot side temper-
ature of 250 F was often reached and it is felt that this should be regarded
as the extremeo The use of the differential thermocouple involved placing
one junction on the hot side and the other on the cold side. The only other
probe necessary was one adjacent on the hot side for obtaining an absolute
temperaturewith the Leeds-Northrup equipment. The "U" value is computed by
measuring the heat flow across the specimen for a given period. This resul-
ted in an "U" factor of 0.073 BTU/(hr) (sq0ft) (°F) or an R of 13.7 for a
waste panel as shown in Figure 2-8 without additional insulation.
Wall Manufacturing Technique—The configuration evaluated, Figures 2-5 and
2-6, was selected for the purpose of ease of production. This simple shape
permits either casting or extrusion. For the purpose of this program
casting was selected for economy reasons. During the course of production,
experimentation with percentage of water was conducted to determine the
effect of a dry viscous mix on strength as against a wet liquid mix. The
ultimate strength was unaffected, only the drying time was greater with
more water. The minimum amount of water required to react the plaster is
prescribed in Table 2-1. The 90 minute plaster was used in this study to
permit sufficient time for manual casting0 Even with this plaster the part
was handleable in 15 to 20 minutes. Limited experiments would indicate that
normal plaster would permit handling in 20 to 5 minutes. This then opens
two attractive possibilities for production.
The first and most readily available approach would be the utilization of
the continuous concrete beam casting system. There are a multiplicity of
these systems available and in use in the United States and elsewhere in
the world. Any system that casts concrete can cast rice hull foam. The
part should be cast on a tool reverse to that shown in Figure 2-7 with the
tool surface textured to the desired surface characteristics of the part.
This would eliminate the need for exterior finishing. The prototype
production was done this way in order to control the critical dimension of
the leg to the interior bond surface. This dimension can be controlled in
110
-------
Heater
Heat Damper
Heat
Distribution
Plate
Specimen
Insulation
Box
Figure 2-29. Calorimeter.
Ill
-------
a continous casting process by a sizing roller following the cast into the
mold. The primary advantage of this process is that no machinery need to
be developed,, The only development costs would be the tooling and a limited
process scale-up study.
In this process the exterior coating of magnesium oxy-chloride can be applied
to the mold surface just prior to that of the rice hull foam. This permits
dimensional stability as both cure at the same time.
The second approach would be the extrusion of the material. A dry mix would
be forced through dies to form the shape onto a conveyor belt running at the
same speed as the extrusion. The material would be cut and trimmed while
still damp. The advantage of this process would be to accelerate production
in a smaller environment. The second approach requires that the coating be
applies separately from the extrusion. This is accomplished by first apply-
ing a thin seal coat without sand. This seals the surface and then the sand
coat is applied. This approach reduces distortion. The disadvantage is that
the equipment and process have to be developed. Once the elements shown in
Figures 2-5 and 2-6 are available, the walls are assembled as shown in
Figure 2-8. The recommended construction approach would be to stand the
elements on one side of the wall onto the sill plate, nail the exterior skin
to the sill plate and use the inorganic adhesive to bond the flanges of the
element together. This would be followed by erecting the elements on the
opposite side against the one completed and bonding the flanges and the two
elements together as shown in Figure 2-8.
Wall Panel Application and Economics—The waste panel is a load-carrying
wall. It has a two-hour fire rating which makes it an excellent partition
wall for multiple family units. Table 2-4 presents the, cost of such a
partition wall installed. In order to provide the full two hours each
exposed surface must be coated with magnesium oxy-chloride. This material
acts as a weatherseal also although not required for this application. The
cost of this material per square foot per surface is $0.111 as presented
in Table 2-5. This makes a total cost installed of $0.922 per square foot.
A conventional two-hour, fire-rated wall costs nearly twice that - a cost
of $1.802 per square foot, see Table 2-6 for details.
The waste wall panel can also be used as an interior panel. Although load
supporting ability and fire protection are not usually required, the avail-
ability of it is an added safety measure. The cost of the waste wall,
$0.700 per square foot installed, is very competitive with the conventional
interior wall cost of $0.76 per square foot, see Table 2-7 for details.
The waste wall panel can also be effectively used as an exterior wall.
Although the load support is required, generally a 1/2 hour fire rating
is considered adequate. A two-hour fire rating is definitely a positive
safety factor. -The exterior wall would utilize a cost of magnesium oxy-
chloride on the weather exposed surface to act as a weather seal. This
would result in a waste exterior wall cost of $0.811 per square foot.
The conventional low cost exterior wall with one half hour fire rating
would cost nearly 40% more or $1.137 per square foot installed (Table 2-8).
112
-------
TABLE 2-4 COST OF WASTE PRODUCT TWO-HOUR, FIRE-RATED PARTITION WALL
Item
Rice Hull Foam 9.656 Ibs/ft ($0.038/lb)
Glass Strands in Cap 0.031 Ibs/ft2
($0.35/lb
Adhesive 0.5 lb/ft2 (0.103)
Assemble on Site:
Four Sections (32 ft2)
Put in Place 16 min
Apply Adhesive 8 min
24 min
Material
Cost
$0.367
.011
.052
11.52/hr
6TH
Seal and Finish Joints
Cost per ft2 #
32 ft2
.020
.450
Labor
Cost
$0.076 *
Included
above
Included
above
Total
Cost
$0.433
.011
.052
.144
.030
.250
.144
.050
.700
Estimate includes factory overhead, depreciation and labor and is based
on MSC continuous production technology.
+ Estimate made from observing demonstration model construction.
* Estimate from Current Construction Costs, 1976 - Lee Saylor, Inc.
* Weather seal and upper and lower plates not included.
113
-------
TABLE 2-5 MAGNESIUM OXY-CHLORIDE COATING COST/SQUARE FOOT
Item Weight,Ibs Cost/lb Cost
MgCl 6H 0 0.36 $0.071 $0.026
^ 2 2
MgO .36 .135 .049
Sand -36 .010 .036
Cost/ft2 - - .111
TABLE 2-6 COST OF A CONVENTIONAL TWO-HOUR,FIRE-RATED PARTITION *
Item Material Labor Total
Cost Cost Cost
Double Stud Wall @ 16" O.C. $0.146 $0.236 $0.382
Two Layers 1/2 in Fire X Sht. Rock .680 .680 1.360
Both Sides
Nails, Clips, and Misc Hardware .060 - .060
Cost/ft2 + .886 .916 1.802
Current Construction Costs, 1976 - Lee Saylor, Inc.
Upper and lower plates not included.
114
-------
TABLE 2-7 COST OF A CONVENTIONAL INTERIOR WALL *
(1/2 HOUR FIRE RATING)
Item Material Labor Total
Cost Cost Cost
Stud Wall @ 16" O.C. $0.073 $0.118 $0.191
One Layer of 1/2 in Sheet Rock .220 .320 .540
on either side
Nails, Clips, and Misc Hardware .030 - .030
+ Upper and lower plates not included.
TABLE 2-8 COST OF A CONVENTIONAL EXTERIOR WALL *
Cost/ft2 + .323 .438 .761
* Current Construction Costs, 1976 - Lee Saylor, Inc.
(1/2 HOUR FIRE RATING)
Item Material Labor Total
Cost Cost Cost
Stud Wall @ 16" O.C. $0.073 $0.118 $0.191
One Layer of 1/2 Inch Sheet Rock .110 .160 .270
one side
One Layer of Exterior Plywood .142 .244 .386
one side
Textured Coating .180 .080 .260
Nails, Clips, and Misc. Hardware .030 ^_ .030
.535 .602 1.137
Current Construction Costs, 1976 - Lee Saylor, Inc.
115
-------
Recommendations and Conclusions— The waste wall panel can be used competit-
ively with conventional construction as a partition wall for multi-family
units, as an interior panel and as an exterior panel.
The two-hour fire rating is a definite bonus for the latter two applications.
The major unknown factor is the effects of long-term loading. It is
recommended that a Phase IV activity be initiated which will permit the
evaluation of this phenomenon over a period of at least 3 years in three
different environments.
116
-------
SECTION 4
ROOF AND CEILING SYSTEM
Material. Selection—The same rice hull foam used for the wall was selected
(Table 2-1). First, it appeared to be the best of the formulation presented.
Secondly, from a manufacturing approach, it is desirable to maintain the same
formulation for as many products as possible.
Design Selection—For the purpose of this study the same design as the wall
was selected. The time and funds available for this study precluded evalu-
ating a multiple variation of configurations because each configuration
required a new tool and a complete series of full-scale tests.
Flexural Strength of Roof and Ceiling System—Although the tensile strength of
the material is low, it was decided to fabricate the first test specimen similar
to the wall without any additional reinforcement. The specimen was built per
Figure 2-30 and was tested in bending as shown in Figure 2-31 with pin-ended
supports. Concrete blocks were placed uniformly on the surface. Blocks were
used from the same production batch and weighed 25 - 1/4 pounds each. The de-
sign load for roofs is 20 pounds per square foot and for floors 40 pounds per
square foot. The major design criterion is that deflection does not exceed the
span divided by 360, (1/360), when at design load. Maximum permissible deflec-
tion for an 8-foot span is then 0.267 inches. The first panel tested failed in
lower skin tension at 28.4 pounds per square foot with a maximum deflection of
0.170 inches. The results are shown in Figure 2-32t
From the results of this test, it was decided to reinforce the lower skin with
one layer of 1/2 inch square mesh hardware cloth (Hail Screen). This improved
the strength of the panel considerably. The panel failed at a load of 105 pounds
per square foot. The deflection at 20 pounds per square foot was 0.055 inches
and at 40 pounds per square foot 0.108 inches, both considerably less than the
allowed 0.267 inches. This would indicate that a reduction in cross-section or
an increase in span length would be acceptable.
Long-Term Loading on Roof Panel—The next aspect to be considered for a floor
or ceiling system is the effects of long-term loading on the panel. The panel
was loaded at a level of 21.9 pounds per square foot. The panel was supported
on blocks as shown in Figure 2-33 to more closely simulate the end conditions
that exist in a housing system. The beam maintained this loading over a period
of 68 days with readings taken as indicated in Figure 2-34. The 20 pounds per
square foot loading condition for a roof is a combination of loads that may
occur in a specific region. These include combination of wind and snow. The
21.9 pounds per square foot loading for a period of 68 days is a rather severe
117
-------
It
1!
il
II
H
II
H
H
II
II
M
II
II
il
H
u
H
u
Hi
M
II
1 1
1 '
I)
i|
il
.'I
|l
il
II
11
U
ll
il
li
n
H
M
ji.
it
n
it
n
i i
M
i i
) !
1'
• I
II
II
il
il
il
II
ll
II
il
i|
;i
"IT '
H
il
u
li
ii
| |
i 1
if
ii
il
il
I'
i '
jl
. |
.1
|T
ll
II
li
II
II
ii
II
li
II
M
ll
|l
I
1 1
|i
1'
1 f
I I
-T
II
ll
il
(1
ll
II
ll
II
il
il
ll
II
il
II
il
1 '
ii
_- .. )(
il
il
t
i;
n
ii
il
II
II
II
t|
P|
II
il
' 1
. . Il
96"
24" —
Figure 2-30. Floor and ceiling specimen.
118
-------
Weight
Panel
L
96"
Figure 2-31. Beam-bending test.
-------
Failure 105 Ibs/ft
•M
O
0
fo
ft)
&
(0
1
O
90
80
70
60
50
40
30
2C
10
With
Reinforcement
Failure 28.4 IfoS/ft2
Without
Reinforcement
0.10
0.20
0.30
Deflection-Inches
1/360
Figure 2-32. Beam-bending vs deflection.
120
-------
Weight
— Panel
Figure 2-33. Beam-bending test (long-term loading)
-------
01
c
o
•H
-P
0)
Q
0.70-
0.60J
0.50.,
0.40-
0.30..
0.20.
0.10.
-
Q-
10
20
30
40
50
60
70
Days Under Load
Figure 2-34. Long-term loading test: Uniformly loaded 21.9 lbs/ft2 beam bending dry.
-------
evaluation. The primary purpose here was to determine the creep characteris-
tics of the material. As can be seen from Figure 2-34, the deformation rate
was as follows:
in/day
0.020
6-39 0.007
39 - 68 0.002
The rate came close to leveling out after 39 days.
Removal of the load indicated an initial set of 0.220 inches. After a period
of 6 days there were no changes.
The results of this study indicate that this panel would be acceptable in
non-snow climates where the maximum load is intermittent. However, in a
snow climate additional reinforcement would be required.
Effects of Water on Long-Term Loading—It was decided to run an equivalent
test with the loaded panel subjected to a complete inundation of running
water. The panel was tested as produced without surface coats. The results
were that the bond at "A", Figure 2030, deteriorated losing the beam effect
and placing the lower skin in pure tension causing failure in 24 hours.
This has resulted in a study of the effect of water on the waste product's
bond.
This long-term test was severe but was designed to provide a rapid evaluation
of the suitability of the waste materials for structures. A load level of
10 to 13 pounds per square foot would be more realistic and the rate of water
flow should be sharply reduced.
A review of the joints of the previous specimen tested under running water
indicated a failure of the substrate below the inorganic adhesive. It was
felt that an epoxy adhesive would penetrate the substrate further and provide
a better joint.
A panel was then manufactured with epoxy bonded joints and again without
surface coating. A load of 10.9 pounds per square foot was determined to
be more near a realistic loading. Figure 2-35 shows the results of this
test to date. The panel showed a decrease in deflection rate after four
days. At the end of 5 days the water was turned off and the panel unloaded
to determine the extent of permanent deformation. After four days of no
load or water the permanent deflection appeared to be 0.283 inches. However,
the panel was still damp and more recovery may have been possible.
\
It was decided to reload the specimen and start the running water. The slope
of the deflection curve the second time appears to be slightly more than that
of the first time.
After three days, the load was removed and the water turned off. The panel
did not recover this time immediately. After three days, it recovered to a
123
-------
CO
Q)
c
o
•H
-p
u
0)
-------
deformation of 0.40 inches and then stablized with no further recovery.
Surface coatings would assist the material appreciably and further studies
should be conducted on the protection of these panels from running water.
Roof Panel Manufacturing Technique—The same techniques recommended for the
wall would be applicable for the roof panel. The reinforcing is an addition-
al complication. If an extrusion process is used, the reinforcement would
have' to be inserted in advance of the die. The hardware cloth is provided
in rolls which permit ease of feeding. If metal lath is used, it is rigid
and would have to be pre-cut to length and inserted incrementally into the
dies.
If a casting process is used, the reinforcement would be installed into the
mold prior to the casting of the rice hull foam.
Roof Panel Product Application and Economics—The major application explored
here was for a load supporting roof with the lower surface used as the ceil-
ing. The assembly technique would be to install all the lower reinforced
sections on the roof first. Then the upper non-reinforced sections would
be bonded to the lower reinforced section creating a structure similar to
that shown in Figure 2-8.
The economics of the waste roof panel as a roof/ceiling combination are
presented in Table 2-9. This compares favorably with the cost of conven-
tional construction presented in Table 2-10.
The waste roof panel, if used as a ceiling would use only the lower elements,
Figure 2-5,' bonded together. Since there are no loads involved other than
dead load, no reinforcement would be required in the elements. The economics
of such a waste system are presented in Table 2-11. This compares favorably
with the cost of conventional construction presented in Table 2-12.
Recommendation and Conclusions for Roof and Ceiling Systems—The rice hull
foam creates an acceptable roof-ceiling combination system, as well as an
acceptable ceiling system. The major concern is the effects of long term
loading with and without water on the product. As can be seen in Figure
2-36, the panel under a short duration load is more than adequate when the
tension side is reinforced. However, the same panelloaded under a typical
21.9 pound per square foot roof load showed considerable creep for the first
6 days and then showed a tendency to stablize although not completely. A
loading condition of this magnitude for such a long period is highly im-
probable. However, the creep characteristics of the material in tension has
been indicated. Improved reinforcement is required.
Running a similar long-term loading with the addition of running water over
the sample indicated further concern. It is doubtful that water would be
be running over an unprotected panel for five days but the test does indi-
cate some sensitivity to the effects of water.
125
-------
TABLE 2-9 COST OF WASTE PRODUCT ROOF-CEILING PANEL WITH HARDWARE CLOTH
AND METAL LATHE REINFORCEMENT
Item
A. Hardware Cloth Reinforcement:
Rice Hull Foam 9.656 lbs/ft2 ($0.38/lb)
Glass Strands in Cap 0.31 lbs/ft2 ($0.35/lb)
Hardware Cloth 0.5 ft2 ($0.1783/ft2)
Adhesive 0.5 lb/ft2 (0.103)
Material
Cost
$0.367
.011
.089
.052
Labor
Cost
$0.076
Included
above
Included
above
Included
Total
Cost
$0.443
.011
.089
.052
9
Assemble Four Sections (32 ft ) on Site
Put in Place 16 min
Apply Adhesive 8 min
above
24 min
11.52/hr
60
32 sqft"
Seal and Finish Joints
Cost/ft2
.020
.539
.144
.030
.250
.144
.050
.789
B. Metal Lathe Reinforcement;
A Above without Reinforcement
Lathe Reinforcement o.5 ft2 ($0.14/ft2)
Cost/ft2
.450
.007
.457
.250
.250
.700
.007
.707
Estimate includes factory overhead, depreciation, and labor and is
based on MSC continuous production technology.
Estimate made from observing demonstration model construction.
Weather seal not included.
126
-------
TABLE 2-10 COST OF CONVENTIONAL ROOF-CEILING SYSTEM
*
Item
2x4 Joists
Plywood Exterior
Sheet Rock, Hang, Tape & Texture
Cost/ft2 +
Material
Cost
$0*126
.142
= 120
.388
Labor
Cost
$0.188
.149
.160
.497
Total
Cost
$0,314
.291
.280
.885
Estimates from Current Construction Costs, 1976 - Lee Saylor, Inc,
Weather seal not included.
TABLE 2-11 COST OF WASTE PRODUCT CEILING PANEL
Item
Rice Hull Foam 4.828 Ibs/ft
Material
Cost
2($0.038/lb) $0.183
Glass Strands in Cap 0.016 lb/ft2 ($0.35/lb) .006
Adhesive 0.25 lb/ft2 ($0.103) .026
Assemble Four Sections (32
Put in Place 16 min
Apply Adhesive 8 min
24 min
Finish Joints
Cost/ft2
ft2) on Site
11.52/hr ,„ .2 _
60 32 ft
.010
.225
Labor
Cost
$0.038 *
Included
above
.144 +
.015 *
.197
Total
Cost
$0.221
.006
.026
.144
.025
.422
Estimate includes factory overhead, depreciation, and labor and
is based on MSC continuous production technology.
Estimate made from observing demonstration model construction.
Estimate from Current Construction Costs, 1976 - Lee Saylor, Inc.
127
-------
TABLE 2-12 COST OF CONVENTIONAL CEILING SYSTEM
* * *
Item Material Labor Total
Cost Cost Cost
2x4 Joists $0.126 $0.188 $0.3X4
Sheet Rock, Hang, Tape & Texture .120 .160 .280
.246 .348 .594
* Estimates from Current Construction Costs, 1976 - Lee Saylor. Inc.
In summary, the waste panel can be used for a roof-ceiling combination system.
However, it should use the steel lathe reinforcement rather than the hardware
cloth. It is not only less expensive but will provide more stiffness. In
addition, it is suggested that reinforcement be incorporated into the comp-
ression side as well as the tension side. This would reduce the potential
for creep. .A weather seal such as a composition roof cover, a built up roof
cover, or any conventional weather seal system should be applied to the ex-
posed surface of the panel. The ceiling system can be used as is for a non-
loaded system.
128
-------
-Hh-
3/8"
to
Steel Lath
2'-0"
5 1/2"
T
1/2"
Figure 2-36. Elements for a floor panel.
-------
SECTION 5
FLOOR SYSTEM
•i .
FLOOR SYSTEM
Material Selection—The same rice hull foam used for the wall and roof was
selected (see Table 2-1).
Design Selection— The floor performs similar to the roof only under high
loads. Therefore, the same design was selected.
Flexural Strength of Floor Panel-—The tests run for the roof were load exten-
ded to include floor requirements. Figure 2-32 shows that the floor is more
than adequate for short duration 40 pound per square foot floor loads. Under
this loading condition an even longer floor span could be considered.
Long Term Floor Panel Loading-The long term loading effects of the panel
used as a floor can be evaulated from Figure 2-34. This test definitely
indicates a problem with the present design. It is therefore proposed to
increase the element depth to 5 1/2 inches to provide additional stiffness.
This will increase the stiffness by a factor of 2.4. This also reduces the
tensile load on the elements surface by 1.6. It is also proposed that metal
lath reinforcement be placed in both surfaces. For details see Figure 2-36.
This additional stiffness and reduction in tensile stress should improve
the long term load carrying capability of the material.
Effects of Water on Long Term Loading— There is no doubt that from the test
data shown in Figure 2-35, it will be necessary to protect the material from
long periods of running water. It is recommended that both surfaces be seal-
ed. Water of evaporation will have minimal effect. It is only prolonged
inundation such as an undetected plumbing leak that would be a problem.
Wear Surface—The exposed rice hull foam surface was used as the wear sur-
face in the initial phases of the demonstrated module construction. As in
most foam type materials the rice hull foam does not wear well. It must be
covered with a wear surface. The minimal acceptable surface would be a
tile. However, it would be more preferable to coat f:he surface with one
quarter inch of magnesium oxy-chloride cement. This is an elastic type
material which does not crack easily and also provides the needed water
seal.
Floor Panel Manufacturing Technique—The same techniques as described for
the roof panel are applicable for the floor panel. If the casting process
is used, the magnesium oxide chloride cement surface can be cast ahead of
the rice hull foam permitting a one step production. If the extrusion
process is used the magnesium oxide chloride would have to be cast on the
element as a secondary step.
130
-------
Floor Panel Application and Economics—The primary application for the waste
floor panel would be for a floor-ceiling system. The panel could also be
used for a floor system but would have less economic advantage. The basic
panel produced from elements shown in Figure 2-36 would cost installed
$0.728 per square foot. See Table 2-13 for details. A one quarter inch
magnesium oxy-chloride wear surface would add $0.220 per square foot (see
Table 2-14) increasing the total cost to $0.948 per square foot. This com-
pares extremely well with a conventional floor ceiling system itemized in
Table 2-15.
There is still a competitive edge for the waste panel as a floor panel when
compared with a conventional wood system (see Table 2-16). Their cost
compares well with a concrete slab of $1.10 per square foot.
Recommendation and Conclusions for the Floor-Ceiling or Floor Panel System-—
The system has definite advantages both for economic and fire safety reasons
in this application. However, the long term loading characteristics of the
system have not been completely determined. The same concern as expressed
for the roof application applies here. Certainly the increased panel depth
and reinforcement in both surfaces would help. However, it is still necess-
ary to evaluate the long term problems over at least a six month period and
preferably over a three year period before using the material for a floor-
in application.
131
-------
TABLE 2-13 COST OF WASTE PRODUCT FLOOR-CEILING PANEL WITH METAL LATHE
REINFORCEMENT
Item Material Labor Total
Cost Cost Cost
Rice Hull Foam 10.208 lbs/ft2($0.038/lb)$0.388 $0.076 * $0.464
Glass Strands in Cap 0.031 lbs/ft .011 Included .011
($0.35/lb) above
Lathe Reinforcement 0.5 sq ft($0.14 sqft) .007 - .007
Adhesive 0.5 lb/ft2 (0.103) .052 Included .052
above
Assemble on Site:
Four Sections (32 sq ft)
Put in Place 16 min
Apply adhesive 8 min 11 c7/h +
24 min g* ' 32 sqft .144 .144
Seal and Finish Joints .020 .030 * .050
Cost/ft2 * .478 .250 .728
*
Estimate includes factory overhead, depreciation and labor and is based
on MSC continuous production technology.
Estimate made from observing demonstration model construction,,
* Estimate from Current Construction Costs, 1976 - Lee Saylor, Inc.
ft
Wear Surface not included.
132
-------
Weight, Ibs
0.72
.72
.72
Cost/lb
$0.071
.135
.010
Cost
$0.051
.097
.072
TABLE 2-14 COST OF MAGNESIUM OXY-CHLORIDE FLOOR SURFACE
Item
MgCl26H20
MgO
Sand
Cost/ft2 .220
TABLE 2-15 COST OF CONVENTIONAL FLOOR-CEILING SYSTEM
Item
2x6 Joists
1/2 inch C-D ext. Plywood
1/2 inch Fire X Sheet Rock
Misc. Hangers and Hardware
Cost/ft2 + .522 .581 1.103
* Estimates from Current Construction Costs, 1976 - Lee Saylor, Inc.
+ Wear surface not included.
Material
Cost
$0.200
•wood .142
: Rock .120
•dware . 060
* * *
Labor Total
Cost Cost
$0.290 $0.490
.131 .273
.160 .280
.060
TABLE 2-16 COST OF CONVENTIONAL WOOD FLOOR SYSTEM
Item
Material * Labor * Total
Cost Cost Cost
Same as Table 2-15 Less Sheet Rock $00402 $0.421 $0.823
Vapor Seal .050 -O20 -^_
Cost/ft2 + -452 .441 .893
* Estimates from Current Construction Costs, 1976 - Lee Saylor, Inc.
Wear surface not included.
133
-------
SECTION 6
FIRE DOOR
Material Selection for Core- The initial work on the fire door core was
based on rice hull foam formulation RH-54 developed in Phase II of this
program. This formulation is presented in Table 2-17. This formulation
when used to construct the fire door was found to be difficult to work
because: (1) rapid setting which made preparation and casting of 2.5 ft
of mixture required for a fire door core extremely difficult, (2) a density
which was too high, and (3) cost somewhat over present materials for fire
door cores. When this formulation was used on laboratory size specimens,
the rapid set-time could be readily accomodated.
One other scale-up problem was that a specimen 6 inches by 6 inches was
readily handleable, even wet, once setting occured. However, a full-scale
door core measures 34.5 inches by 77 inches by 1.5 inches. When wet, the
cast material will crack on handling because of the high weight (water
present prior to drying) and long unsupported distance of cast panel.
It was necessary to reformulate so as to accomodate processing, properties
and price requirements. Several approaches were used and these were:
1. Reduce density by use of styrofoam beads
2. Increase processability by use of beads because of processability and
ease of movement to fill the mold
3. Achievement of 1 & 2, above, by foaming
Table 2-18 lists a series of formulations which were made and evaluated
in the laboratory. This evaluation was accomplished by exposing each
specimen to a propane torch kept 3 inches from the front surface and
a thermocouple to the rear face. It must be commented that the propane
torch provides a highly erosive action not present in the standard fire
test. The use of comparative results and performance of a "standard"
material made this test useful for screening.
The data showed that direct replacement of rice hulls by styrofoam beads
caused thermal insulation to be unsatisfactory and the most logical reason
being the great distance in density from 40.9 pounds per cubic foot.
134
-------
TABLE 2-17 RH-54 FORMULATION
Rice Hulls lOOg
1" Chopped Glass Fiber 15g
MSC Additive 40g
Gypsum Plaster 60g
Casting Plaster 240g
Water 225g
Formulation #2 has aluminum hydrate added to provide extra water as
possible way to improve insulative performance. This was very effective
without too great an increase in density but large increase in cost. In
Formulation 3 & 4, the aluminum hydrate was replaced by an even more
efficient source of water, aluminum sulfate. This has 48.6% water as
sompared to 34.0% for aluminum hydrate and 21.0% for hydrated gypsum.
The incorporation of aluminum hydrate provided improved insulative perfor-
mance and this was due to the ability of this material to provide water
for transpirational cooling. Assuming this to be correct, new addition of
a compatible material having a higher water content might give still further
improvement. For this purpose, Al (SO ) 18H O was selected.
135
-------
TABLE 2-18 FIRE-DOOR INSULATION MATERIAL FORMULATIONS BASED ON RICE HULLS
Formulation
Item 12345
Water 330 350 330 340 360
90 min. Casting Plaster 450 450 450 450 450
MSC Proprietary Binder 60 60 60 60
Styrofoam Beads 55 - -
Rice Hulls 75 75 150 150 150
M2°3 3H2° - 164 - -
Al (SO } 18H 0 - - 50 * 50 * 50
£, ^x -J £,
1/2" Chopped Glass - - - - 3.5
G-3300 (Surfactant) - - - - 3.5
6 7
350 360
450 450
-
5
75 150
164 100
50
3.5
3.5 3.5
Density, Ibs/ft 20.5 26.2 37.3 26.4 27.4 28.6 28.6
Cost, $/ft3 0.875 1.517 1.712 1.138 1.915 1.208 1.739
Max. Temp, °F after 2 hrs Failed £300 225 Failed $ 404 475 235
Formulation
Water
90 min. Casting Plaster
MSC Proprietary Binder
Styrofoam Beads
Rice Hulls
Al O 3H O
1/2" Chopped Glass
G-3300 (Surfactant)
Density, Ibs/ft
Cost, $/ft
Max Temp, F after :
Failed +
See notes on following page.
8
360
450
150
50
3.5
3.5
28.5
1.038
rs340
9
360
450
150
-
50
-
3.5
27.2
0.584,
335
10
350
450
150
-
50
3.5
3.5
33.9
0 .^940
155
11 #
425 *
450
12.5
-
12.5
3.5
3.5
33.0
0.796
185
12 *
425 ff
450
12.5
38
12.5
3.5
3.5
33.3
0.964
250
13
350
450
150
-
50
3.5
3.5
33.0
0.958
300
14
330
450
15
150
-
50
3.5
-
26.8
0.919
285
136
-------
TABLE 2-18 (Continued) FIRE-DOOR INSULATION MATERIAL
BASED ON RICE HULLS
Added a solid: All following formulations use 33% solution,
thus 150 g sol'n contains 50G Al2(804)318H20.
+ Reached 600 P in 28 mins with rapidly increasing rate of heat-up.
^ Reached 435 F in 20 mins and rising very rapidly.
Water added to make up for that lost because of less amount of
Al2(SO4)3l8H2O solution used.
137
-------
Addition of this compound to a water-plaster mixture caused setting to occur
in about 15 minutes even though 90 minute casting plaster was used. What
was unanticipated was a 75% increase in volume due to frothing. It appears
as if other additions are needed to prevent cracking or drying of the foam.
The glass fibers and/or rice hulls used in the formulations should prevent
cracking.
The same frothing and rapid set was obtained when regular wall plaster was
used.
Frothing can not occur between aluminum sulfate and calcium sulfate. The
presence of calcium carbonate in the plaster reacts with the acidic aluminum
sulfate to generate carbon dioxide, more calcium sulfate and aluminum
hydroxide (alumina trihydrate). So use of aluminum sulfate adds water for
cooling and forms two other materials which also help in insulation because
of water content.
Formulation 5 was adequate for 2 hour use, with a large decrease in cost
but with very satisfactory density. Formulation 6 and 7 are too expensive
but Formulations 8 thru 14 all are acceptable.
It was decided that further evaluation had to be made on full scale door
cores. Initially it was decided to case the core into an upright mold.
Modifications to Formulations 5 and 14 from Table 2-18 were used to cast.
five cores. These are presented in Table 2-19.
It was obvious from this work that the panels were not drying on the bottom
and were so wet as to cause fracture when moved, support of underside during
movement might prevent this. Frothing could also cause weakening, so "B"
was made without detergent. There was now insufficient material to fill
the mold. Slight increase in detergent gave better results for formulations
C, D, and E but results were still not satisfactory.
It was decided to do further refinement on Formulation E with a specimen
24 inches by 32 inches. This size provided scale up data, permitted E-119
fire test evaluation and used less time and materials.
The processing difficulties from rapid set time were overcome by use of a
protein retarder and it became possible to fabricate specimens having a
material density of about 28-32 Ib/ft . An E-119 fire test in which face
temperatures were increased (by increment) to 1800°F demonstrated complete
protection for a period of 90 minutes as evidenced by failure of cotton
linters or Kleenex to ignite when pressed agianst the back surface during
the test.
The maximum recorded back-side temperature was 340°F. After 90 minutes,
the test specimen was placed at a 20 foot distance from a fire hose with
a 105 inch nozzle and hit with a stream of water at 30 psi for 8 seconds;
the door maintained its integrity despite the erosion of material to a
depth of about 1.0 inch.
The above evaluation resulted in the selection of the formulation presented
in Table 2-20.
138
-------
TABLE 2-19 ATTEMPTS TO CASE FULL SIZE FIRE DOOR CORE (34.5" x 77" x 1.5")
Constituent
Rice Hulls
A
8978
Formulation
BCD
8978 8978 9427
E
9427
1/2" Glass 209 209
90 min. Casting Plaster 26933 26933
Water 20948 20948
Al (SO ) 18H 0 Sol'n 8978 8978
^ *X *J ^
Detergent 80 0
209 219.5 219.5
26933 28280 28280
25436 26708 26708
2244 2356 2356
25 26.3 26.3
TABLE 2-20 FORMULATION FOR FIRE DOOR CORE
Item %
Rice Hulls
1/2 inch Chopped Glass Fiber
90 min Casting Plaster
MSC Binder
Water
Al (S04)
Detergen
Retarder
Density - 32 pounds per cubic foot
Stablized Moisture Content - 10%
13.7
0.6
41.1
3.13
38.0
3.4
0.04
0.03
139
-------
Material Selection for Framing Members—The framing material has screw
retention requirements as well as fire resistance requirements. The density
of this material is no longer critical since the volume of material needed
for a frame is small compared to the core.
The first experiments were carried out to compare plaster, magnesium oxy-
chloride and magnesium oxy-sulfate with phenolic microballoons which impart
lower density and helps nail and screw retention. Results are summarized in
Table 2-21.
These data indicated the weakness of plaster and superiority of magnesium
oxy-chloride.
The three binder materials were then compared to a fibrous material prepared
from Douglas Fir bark, called Silvacon, assuming this would help values using
plaster. Results are presented in Table 2-220 Use of Silvacon improved the
properties of plaster over phenolic microballons but are still inadequate for
the application. The magnesium oxy-chloride again displayed far superior
results and adequate properties for framing material.
It was decided to continue formulations with magnesium oxy-chloride in place
of plaster but to investigate other filler materials derived from waste
materials. Since Silvacon is produced from a scrap material (Douglas Fir
Bark) and is inexpensive, it was included in the study.
A variety of formulations were evaluated with the various waste materials and
magnesium oxy-chloride. Results are summarized in Table 2-23. The properties
that resulted from use of Douglas Fir fibers gave excellent results even when
the bulk of this material was replaced by waste-derived Silvacon (see Formula-
tion 1). Actually this mixture of fillers gave highest all around strengths.
Since the use of Douglas Fir fibers was so successful it was tried with
plaster, Formulation 5, and gave poor properties. Formulations 3 and 4 were
made with 90 mesh sand added to increase density since nail and screw retention
improved with density but with plaster this improvement fell far too short.
Magnesium oxy-chloride and milled rice hulls, Formulation 6, provide nearly
an equivalent product to Formulation 1.
Mechanical tests were conducted on wood materials used for frames. These
results are presented in Table 2-24. As can be seen the 455 pound screw
retention capability of inorganic mix #1 and the 445 screw retention capabil-
ity of inorganic mix #6 are as good as or superior to that of the wood.
One additional problem was noted when eight foot long sections were fabricated:
low flexural strength. To improve this characteristic untwisted sisal strands
were incorporated into the mix. This resulted in the selection of the material
formulation presented in Table 2-25.
140
-------
TABLE 2-21 COMPARISON OF MATRIX MATERIALS FOR DOOR FRAMING
USING PHENOLIC
Item
MgCl26H20
H20
MgO
MSC Binder
MgS047H20
Casting Plaster
Phenolic Microballons
Density, Ib/ft
Nail Pull, Ibs
Screw Pull, Ibs
MICROBALLOONS
1
93.5 g
7005 g
100.0 g
10.0 g
-
-
30.0 g
56
Not nailable
235
TABLE 2-22 COMPARISON OF MATRIX MATERIALS FOR
USING SILVACON
Item
MgCl26H20
H20
MgO
MgSO 7H 0
MSC Binder
Casting Plaster
Silvacon
Density, Ib/ft
Nail Pull, Ibs
Screw Pull, Ibs
CoTtroression. nsi
1
93.5 g
80.5 g
100.0 g
-
15oO g
-
75.0 g
82
395
435
3100
Formulation
2
-
95 g
100 g
-
66 g
-
30 g
59
22
63
DOOR FRAMING
Formulation
2
-
95
100 g
66 g
-
-
75 g
81
Not nailable
88
1350
3
-
175 g
-
-
-
275 g
30 g
58
Not nail
able
32
3
-
175 g
-
-
-
275 g
75 g
61
29
94
141
-------
TABLE 2-23 USE OF VARIOUS WASTE FILLERS FOR DOOR FRAMING USING
MAGNESIUM OXY-CLORIDE
Item 1
MgCl 6H 0 93.5 g
^ Cf
HO 70.5
MgO 100.0
Regular Rice Hulls -
2
93.5 g
70.5
100.0
35.0
Formulation
3 4
-
175 g 175 g
-
- -
5 6
93.5 g
175 g 70.5
100.0
-
=5 Milled Rice
Hulls *
Douglas Fir 25.0 25.0
Fibers
Silvacon 50.0
Sand
MSC Binder - 10.0
Casting Plaster
Density, Ib/ft 74 69.5
Nail Retention 225 200.0
Ibs
Screw Retention 455 360.0
Ibs
Compression, psi 3250 1350.0
100
500
60 g
60 g
125 g 125 g
275 g 275 g
74 86
15 65
130
500
50 g
275 g
69.5
35.0
85.0
530.0
40.0
40.0
81.0
445.0
2400.0
Hammer-milled rice hulls, contains about 33% (wt) of - 20 mesh fines.
From Industrial Paper Company, Longview, Washington. Screened to
remove -20 mesh fines.
142
-------
TABLE 2-24 PROPERTIES OF VARIOUS WOODS USED IN DOOR FRAMES
Sample
Fir
Alder
Hemlock
Maple
d
0.32
0.50
0.59
0.70
%H2°
gain
118.1
71.9
44.3
64.8
Compression,
psi
Dry
4890
5235
9328
7910
Wet
1919
3607
3432
4068
Screw Nail
Retention Retention
Ibs
Ibs
Dry Wet * Dry Wet
263 125 93 88
415 265 247 128
348 195 100 *
495 + 325
One week submersion in water.
+ Chemically treated for use in fire door frames.
^ One test, other specimen failed to tolerate nail.
Both specimens failed to tolerate nail.
*
#
#
TABLE 2-25 FORMULATION FOR FRAMING MATERIAL
Constituent %
MgCL2 6H20
H20
MgO
MSC Binder
Douglas Fir Fibers
Silvacon 412
Untwisted Sisal
26
19
28
2
6
17
2
143
-------
Design Selection—The design options on a fire door are limited. The core
is limited to one and one half inches for a 90 minute fire rating. The fram-
ing materials are also limited to that of conventional design. A cross-section
of the door is shown in Figure 2-37. The primary purpose of the groove in
the stiles and rails is to element a direct heat or flame path through the
wallo
Mechanical Properties —The mechanical requirements for the core are to be
light as possible and capable of being handled. The core when completely
dried weighs 32 pounds per cubic foot. Existing cores weigh between 20
and 35 pounds per cubic foot which places the waste core in the higher density
classification.
The mechanical requirement of the framing material is screw retention. The
screw retention of waste framing material is 400 to 450 pounds. This compares
favorably with that of conventional wood framing materials whose retention is
from 260 pounds to 495 pounds (see Table 2-15).
Fire Test — The fire test for the door follows the ASTM E-152 procedure. The
time versus temperature requirement is the same as the ASTM E-119 curve shown
in Figure 2-18. The difference in procedures is that the door is installed
in a wall. The assembly is placed in a furnace for 90 minutes and then
removed and sprayed with a 30 psi hose stream for a period of seconds equal
to 0.9 times the surface area of the door in square feet.
Preliminary tests were conducted on panels measuring 24 inches by 32 inches
by 1 1/2 inches, exposed to a furnace programmed to the ASTM time-temperature
curve. Temperature on the back side was measured by thermocouple as was the
flame temperature. A small wad of cotton and/or tissue was also kept in
contact with the back surface.
After 90 minute exposure the backside temperature was insufficient to ignite
the tissue. The specimen was moved to where it was exposed to a stream
of water at 30 psi from a 1.5 inch nozzle at a distance of 20 feet. After
8 seconds of such exposure, the sample remained intact. Since the cotton
failed to ignite and the back panel was in excellent condition and the water
did not wash away all of the core, the test was considered to be very success-
ful (see Figure 2-38).
The results of these tests were sufficient to encourage a full-scale test.
Two sets of cores and framing materials were produced under the observation
of Warnock Hersey International, Inc. who will conduct the certification
tests under the next phase. These elements were shipped to Cal Wood Door
for fabrication into a test assembly.
Core Manufacturing Techniques— The core material is cast into a mold and
the mold inserted into a matched platen pressure. The mold is closed to
its stops and a pressure of one to two pounds per square inch maintained to
equalize pressures generated by the foaming action.
144
-------
Wood Facing
Core
1 1/2'
Cn
Stiles or
Rails
1"
CORE
Figure 2-37. Fire door design.
-------
Figure 2-38. Surface after 90-min fire and hose stream.
146
-------
Framing Material Manufacturing Techniques — The elements for this study
were made in matched molds in a press above. However, the consistency of the
compound is such that an extrusion process would be the logical procedure
for large quantity production.
Fire Door Product Application and Economics— The application explored here
is the 90 minute fire door. Although there are other cores competing for
this market, the only framing material is chemically treated wood which is
marginal* The economics of the core are presented in Table 2-26. The material
cost of the core is $2.038 for a three foot wide, six foot eight inch high
door. Labor costs to produce it would be approximately $1.00 per door.
Competitive cores are provided the door manufacturer in a price range of
$10.00 to $15.00 per unit. When marketing, distribution and waste expenses
are added to the waste door, it would appear to be competitive in that
market.
*
The framing material cost in Table 2-27 of $0.0572 per linear foot is
commercially competitive. Furthermore the material is fire resistand and
will maintain screw holding power easily for 90 minutes of fire exposure.
The core and framing material can both be utilized in the door industry. It
is necessary to generate an interest by a door manufacturer in order to
productize this material. This has been accomplished. During Phase IV it
is proposed to guide and direct these manufacturers in order to encourage
production of doors from waste material.
147
-------
TABLE 2-26 FIRE DOOR CORE ECONOMICS
Volume
Dry Weight
Wet Weight
1.896 cubic feet
60.67 pounds
79.55 pounds
Item Ibs
Cost/lb
Cost
Rice Hulls
1/2 inch chopped glass fibers
Casting Plaster
MSC Binder
Water
Al (S04) 318H20
Detergent
Retarder
10.9
s .5
32.7
2.5
3002
2.7
= 03
.02
$0.062
.40
.0252
.010
-
.10
.75
.98
$0.676
0.200
0.824
0.025
-
0.270
0.023
0.020
79.55
2.038
Price delivered to factory.
Table 2-27 FRAMING MATERIAL ECONOMICS
Item
MgCL 6H O
^ £,
MgO
MSC Binder
Douglas Fir Fibers
Silvacon 412
Untwisted Sisal
Cost per linear foot
linear foot =0.78 Ibs
Weight, Ibs
0.202
0.148
0.218
0.016
0.047
0.133
0.016
Cost/lbs
$0.071
-
0.135
0.010
0.060
0.075
0.030
Total Cost
$0.0143
O.Q224
O.OOQ2
0.0028
0.0100
0.0005
0.0572
148
-------
SECTION 7
DEMONSTRATION UNIT CONSTRUCTED FROM WASTE MATERIALS
Sheets, panels, and test specimens can define a considerable number of proper-
ties and evaluate materials and systems to a point. The true and complete
evaluation occurs only when these are incorporated into a complete demonstrat-
ion unit. Only when floors, roofs and walls are used as such can the real
applicability be determined. Therefore, a demonstration unit was designed
and built using the previously described waste materials.
It was requested that the unit be disassembleable to permit easy transportation
to various exhibition sites. The initial site was the Environmental Protection
Agency Environmental Research Center in Cincinnati, Ohio. The disassembly
requirement created problems with joints and eliminated weather seal
capability which requires that the module be exhibited indoors only. The
details of the construction are presented in Appendix E.
The unit constructed is shown in Figure 3-1. The inorganic panels had to be
constructed by hand using the techniques previously outlined. Since this
was the first production of any large quantity of these materials considerable
information was developed on the use of the waste material in housing units.
These details are presented below.
ORGANIC WASTE PANELS
Effect of Moisture— The organic waste panels both wood waste and peanut shells
were used as paneling material for both the window and door panels. The
construction was accomplished during an extremely humid period. It was
determined that the wood waste panels absorbed moisture which caused con ider-
able swelling. This situation was corrected by sealing the surfaces with a
urethane coating. The peanut shell material did not demonstrate the same
phenomena.
Applicability— The organic waste panels were workable with conventional tools
and could be nailed without difficulty.
INORGANIC WASTE PANELS
Wall Surface Finish— In production it is proposed that the desired surface
finish be incorporated into the waste wall panels. However, due to the limit-
ed hand-built quantities utilized here it was decised to be more efficient for
this unit to apply the surface finish as a secondary step. Conventional dry
wall spackling compound was used on the interior surfaces and interacted very
well with the waste wall.
149
-------
Figure 3-1. Demonstration unit,
150
-------
Such conventional finished techniques are completely compatible. On the
exterior finish a conventional stucco compound was tried first. Although
it appeared to work well on small specimens, it did not adhere well on the
surface of the module. This material had to be removed.
Laboratory tests showed that magnesium oxy-chloride when mixed with sand
made an excellent surface coat. It was originally used as a stucco material
before the advent of gypsum stucco. Tests showed that the material interacted
with the rice hull foam to affect a good bond. However, on small specimens
the effect and mechanics of this interaction were not completely displayed.
The magnitude was learned only after the exterior of the unit was coated.
An outward curvature towards the magnesium oxy-chloride surface was noted on
all walls when the unit was being dismantled for shipment. An upward
curvature or crown was noted on the floor panel because the wear surface appli-
ed was magnesium oxy-chloride also. However, no curvature was detected in
the roof panels which did not receive a magnesium oxy-chloride coat. To
verify the effect of magnesium oxy-chloride on the rice hull foam, an eight
foot long section of 1/2 inch thick rice hull foam was coated with magnesium
oxy-chloride in the same fashion as the unit and observed. The magnesium
oxy-chloride penetrated to a depth of 1/4 inch causing an expansion of the
rice hull foam which resulted in a curvature towards the coating. Tests
indicate no effect on strength. The test specimen indicated that the expansion
action after 48 hours. To verify that this action was stabilized, the wall
panels were measured over a five day period and no further movement was
detected.
Experiments show that if the magnesium oxy-chloride is cast onto the surface
at the time the rice hull foam is cast, no curvature is generated. It also
appears that if a very thin film of the solution without sand is applied to
the cured surface first a seal is created which eliminates excessive amounts
of absorbtion of the solution by the rice hull foam when the coating is
applied. This eliminates the curvature.
Floor Wear Surface— The floor panels were uncoated initially. However,
after three weeks of technicians working on the interior of the unit, the
foam surface showed signs of considerable wear. It was determined to provide
a wear surface of magnesium oxy-chloride cement. The results of long term
studies discussed in previous sections indicated that the floor panel should
be reinforced in both surfaces. Therefore, it was decided to apply a rein-
forced wear surface of magnesium oxy-chloride cement.
Applicability— The inorganic waste panels were workable with conventional
tools and ordinary construction methods were applicable.
DOORS
Applicability — The doors required no special consideration and were installed
in a similar manner as conventional doors.
151
-------
APPENDICES
APPENDIX A. INFORMATION SOURCES ON THE ECONOMICS OF PARTICLE BOARD
Material Systems Corp. 1975. A Study of the Feasibility of Utilizing Solid
Wastes for Building Materials. Monthly Progress Reports 6023-11 to 14,
ETR 94-95, and Phase II Summary, Escondido, California.
National Bureau of Standards. 1966. Mat-Formed Wood Particleboard (CS236-66).
U.S. Department of Commerce, U.S. Government Printing Office, Washington,
D.C.
USDA Forest Service. 1975. The Outlook for Particleboard Manufactured in the
Northern Rocky Mountain Region, USDA Forest Service General Technical
Report INT-21. Intermountain Forest and Range Experiment Station,
Ogden, Utah.
Vajda, P. 1967. The Economics of Particleboard Manufacture. Presented at
the Particleboard Symposium, Washington State University, 1967.
Vajda, P. 1970. The Economics of Particleboard Manufacture Revisited or an
Assessment of the Industry in 1970. Presented at the Particleboard
Symposium, Washington State University, 1970.
Vajda, P. 1974. Structural Composition Boards in the Wood Products Picture.
Presented at the Particleboard Symposium, Washington State University,
March 1974.
152
-------
APPENDIX B. DATA ON WATER EFFECTS ON RICE HULL REINFORCED INORGANIC MATRIX
TABLE B-l UNCOATED: RICE HULL REINFORCED INORGANIC MATRIX — WEIGHT DATA
Original Weight Weight After Weight After Weight
1 Day 1 Week After 1
Month
1 36.7 - - -
2 37.4 - - -
3 36.8 - -
4 37.4 - 36.2
5 36.8 - 36.3
6 36.0 - 34.8
7 35.7 - - 34.4
8 35.7 - - 34.3
9 37.6 - - 36.1
10 36.6 45.7
11 36.3 45.2
12 37.1 46.1
13 37.3 - 46.8
14 34.9 - 44.8
15 37.5 - 47.9
16 36.5 - - 48-3
17 36.6 - - 48'5
18 35.7 - - 47-!
19 36.9 46.5 - ~
20 37.2 46.2 - ~
2i 37.8 46.9
22 37.3 - 45-6
23 36.2 - 45'7
24 37.0 - 44-l -
25 38.0 - - 39.2
26 38.0 - J9.6
27 36.1 - - 41'8
28 36.8 45.9 -
29 36.8 45.3 - ~
30 36.7 46.1
31 37.9 - 46/8 '
32 37.0 - 46'5
33 38.2 - 46:3
34 35'2 ' '
_ 49.0
" -
36 35.4
35 3"7-1 " - 47.6
153
-------
TABLE B-2 UNCOATED: RICE HULL REINFORCED- INORGANIC MATRIX
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Condition
Set in
open air
No water
Control
Specimens
1 day Static
Water
Submersion
1 Week Static
Water
Submersion
1 month Static
Water
Submersion
1 day in
Water Chg.
Daily
1 Week in
Water Chg.
Daily
1 month in
Water Chg.
Daily
1 Day Static
Water Sub.
Drying Out
1 Week Static
Water Sub
Test After
Dry Out
1 Mo. Static
Water Sub.
Test After
Dry Out
Comp Strength
525
580
575
550
575
500
545
475
570
300
355
300
345
310
340
320
355
300
310
335
340
330
310
370
265
255
320
495
490
490
595
530
525
410
460
465
Avg.
% Ret.
543
318
332
325
328
337
280
492
550
445
58.6%
61.1%
59.8%
60.4%
62%
51.6%
90.5%
101%
81.9%
154
-------
TABLE B-3 COATED ZYNOLITE PAINT: RICE HULL REINFORCED INORGANIC
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
MATRIX (COATED)
ginal Weight
38.6
37.8
37.5
38.6
38.2
38.2
38.3
38.6
38.2
38.0
38.2
38.2
40.5
39.1
38.8
36.1
37.6
38.1
39.1
38.0
38.0
38.8
39.2
38.7
37.9
38.4
38.3
38.4
36.6
38.2
38.5
38.5
38.1
38.5
38.8
39.5
— WEIGHT DATA
Weight After
1 Day
_
-
-
-
-
-
_
_
-
47.4
48.7
48.8
-
_
_
_
_
-
50.2
48.6
47.2
-
_
-
~*
"•"
—
49.0
46.9
48.7
-
—
~
"™"
—
—
Weight After
1 Week
_
-
-
38.5
38.2
38.2
-
-
-
-
-
-
50.6
49.8
49.2
-
-
-
—
—
-
49.3
49.1
48.2
"""
~
™~
49.5
49.5
48.7
Weight
1 Month
-
-
-
-
-
-
38. '3
38.4
38.1
—
—
—
—
—
-
48.7
50.6
51.4
~
~
™
_
~
48.9
50.8
40.6
*
~
50.8
51.9
52.8
155
-------
TABLE B-4 COATED ZYNOLITE PAINT: RICE HULL REINFORCED INORGANIC MATRIX (COATED)
Condition
1 Set in
2 Open Air
3 No Water
4 Control
5 Specimens
6
7
8
9
10 1 Day Static
11 Water
12 Submersion
13 1 Week
Static
14 Water
15 Submersion
16 1 Month
Static
17 Water
18 Submersion
19 1 Day in
20 Water Chg.
21 Daily
22 1 Week in
23 Water Chg
24 Daily
25 1 Month in
26 Water Chg.
27 Daily
28
29 Did not Dry
30 Specimens
31 1 Week Static
Water Sub.
32 Test After
33 Dry Out
34 1 Mo Static
Water Sub.
35 Test After
36 Dry Out
^
Comp Strength
610
510
575
500
560
570
585
600
510
320
335
320
335
-
325
260
265
_
350
305
315
340
350
350
355
350
305
325
375
_
_
_
600
_
600
635
_
490
525
490
Avg.
_
-
-
-
581
-
-
-
_
_
325
-
_
_
307
_
_
__
307
w.
_
335
_
_
352
__ _
_
335
_
_
_
_
612
_
_
501
% Ret.
_
_
_
_
_
—
_
_
_
_
55.9%
_
_
52.8%
52.8%
57.6%
60.5%
57.6%
10.5%
86.3%
156
-------
TABLE B-5 COATED LYTRON 621: RICE HULL REINFORCED INORGANIC
Weight
1 Month
I
2
3
4
5
6
7 36.1 - - 36.2
8 34.8 - - 35.0
9 35.4 - - 35.5
10
11
12
13
14
H « , - - 49.2
35.7
48.3
jt . o
19
20
21
22
23
24 - •"•;- _ - 45.6
25 36.1 _ _ 44>9
26
27
28
29
30
31
32
33
34
35
36
MATRIX
Original
35.7
35.5
36.4
35.5
35.8
35.3
36.1
34.8
35.4
35.6
34.8
35.6
34.9
34.3
35.7
35.1
35.7
34.8
35.9
35.4
36.0
35.8
35.7
34.5
36.1
35.2
35.3
35.2
35.3
35.5
34.4
35.3
35.6
35.4
34.4
35.2
(COATED) — WEIGHT DATA
Weight Weight After
1 Day
_
-
-
-
-
-
-
-
-
46.9
46.5
46.5
-
-
-
—
~
—
47.0
46.2
46.8
-
-
—
™"
~
"""
46.5
45.8
46.5
-
-
-
""*
~
""
Weight After
1 Week
_
-
-
35.7
35.8
35.6
-
-
-
-
-
—
46.3
45.5
46.6
~
_
~"
~
~
—
46.6
46.5
44.7
~
~
46.4
46.3
47.1
157
-------
TABLE B-6 COATED LYTRON 621: RICE HULL REINFORCED INORGANIC MATRIX (COATED)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Condition
Set In
Open Air
No Water
Control
Specimens
1 Day Static
Water
Submersion
1 Week Static
Water
Submersion
1 Month
Static Water
Submersion
1 Day in
Water Chg.
Daily
1 Week in
Water Chg.
Daily
1 Month in
Water Chg.
Daily
Did Not
Dry
Specimens
1 Week Static
Water Sub.
Test After
Dry Out
1 Mo. Static
Water Sub.
Test After
Dry Out
Comp Strength
555
510
500
400
465
485
480
380
475
305
275
280
295
285
370
335
330
315
310
335
315
270
245
285
300
205
265
510
550
545
470
490
445
Avg.
% Ret.
484
287
317
327
320
267
287
535
468
59.2%
65.4%
61.5%
66.1%
55.1%
59.2%
110%
96.7%
158
-------
TABLE B-7 COATED W-1Q2: RICE HULL REINFORCED INORGANIC
MATRIX (COATED)—WEIGHT DATA
Original Weight
Weight After
1 Day
Weight After
1 Week
Weight
After 1
Month
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
38.1
38.1
37.9
38.2
38.1
38.7
37.0
37.4
37.2
38.5
37.5
37.4
38.1
37.4
39.3
38.3
38.8
38.5
38.6
38.0
38.0
38,4
37.8
38.2
36.8
38.1
37.7
37.8
38.0
3707
37.5
38.2
37.9
38.2
37.5
38.2
49.5
48.4
48.2
49.5
48.6
48.8
48.6
48.6
48.2
38.2
38.1
38.6
49.6
48.5
50.8
49.9
48.7
49.6
48.9
49.9
49.5
37.1
37.4
37.2
52.6
53.1
52.6
48.3
49.9
50.0
52.2
52.0
52.7
159
-------
TABLE B-8 COATED W-102: RICE HULL REINFORCED INORGANIC'MATRIX (COATED)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Condition
Set in
Open Air
No Water
Control
Specimens
1 Day Static
Water
Submersion
1 Week Static
Water
Submersion
1 Mo. Static
Water
Submersion
1 Day in
Water Changed
Daily
1 Week in
Water
Changed
Daily
1 Month
in water
Changed
Daily
1 Day Static
Water Sub.
Test After
Dry Out
1 Week Static
Water Sub.
Test After
Dry Out
1 Mo. Static
Water Sub.
Test After
Dry Out
Comp Strength
555
525
540
515
545
555
485
535
490
300
305
265
320
285
315
315
310
310
310
260
260
310
320
260
275
305
300
325
310
350
335
370
310
460
435
495
Avg.
% Ret.
527
290
367
312
277
297
293
328
338
463
463
55%
58.2%
59.1%
5205%
56.3%
55.7%
62.3%
64.2%
87.9%
87.9%
160
-------
TABLE B-9 KILN-DRIED DOUGLAS FIR LUMBER—WEIGHT DATA
Original Weight Weight After Weight After Weight
1 DaY 1 Week 1 Month
1 15.3
2 14.8
3 15.3 _ ~
4 - ~
5 - I
6 - -
7 - _
8 - -
9 - - - _
10 -
11 -
12 -
13 15.5 _ 24.6
14 15.5 - 24.2
15 17.4 _ 26.0
16 15.7 _ _ 29.6
17 15.0 _ _ 28.9
18 15.4 _ _ 29.8
161
-------
TABLE B-10 KILN-DRIED DOUGLAS FIR LUMBER
Condition
Comp Strength
Avg.
% Ret.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Test Dry
Control
Specimen
1 Week Static
Water
Submersion
1 Mo. Static
Water
Submersion
5150
5400
5600
5383
3050
3000
3300
2950
2950
3000
3117
2967
57.5
55.1%
162
-------
PHASE IV SUMMARY REPORT
SECTION 1
INTRODUCTION
Material Systems Corporation and the Solid and Hazardous Waste Research
Division of the Municipal Environmental Research Laboratory at Cincinnati,
Ohio, have conducted a joint development study of the application of waste
materials to residential and commercial construction. This is a summary
report of the Phase IV activity.
Results of previous work have been reported in summary reports for each
phase. Only a brief resume of these results is presented here for the
convenience of the reader to serve as an introduction to the detailed
discussion of Phase IV.
In Phase I, a comprehensive literature search was conducted on potential
matrix, reinforcement and filler candidates. The more promising candidates
were evaluated in laboratory studies. From these studies, two types of matrices
(furfural derivatives and inorganics) were selected for evaluation with the
two matrices.
The Phase II activity demonstrated the potential technical and economic
feasibility of utilizing waste materials for building construction. As
the Phase II study progressed, it became apparent that there existed many
product possibilities from waste materials. Therefore it was jointly decided
to concentrate Phase II activity on product development and to postpone specific
qualifications until Phase III.
The product development effort included testing and evaluation to a
sufficient level to assure probable qualification for specific products.
The products selected for qualification studies in Phase III were in two
main categories: those materials from organic matrices and those from
inorganic matrices.
The Phase III study selected wood waste and peanut shells for possible
qualification as organic particle board. Panels as large as 4 by 4 feet
were fabricated and tested. However, not all the processing problems
were resolved; this would require further studies. The wood waste
particle board probably has some economic advantage over commercial part-
icle boards. The economic advantage of the peanut shell board is marginal.
With the resolution of some of the processing problems, the boards could
probably be qualified. Their principal limitation is high density: boards
from this process have a minimum weight of 55 pounds per cubic foot.
163
-------
The Phase III study also selected a composite made of rice hulls and an
inorganic matrix for use as walls, floors, roof-ceiling, and floor-ceiling
systems. A satisfactory load carrying 2-hour fire-rated wall was developed.
Acceptable floor, floor-ceiling, and roof-ceiling systems were also developed.
These building components appear to offer a cost advantage over conventional
structures and to lend themselves to mass production techniques. Further
studies were recommended on the problem of long-term loading and were
carried out in Phase IV, as described in this report.
The rice-hull foam mentioned above was modified to be used as a core for
fire doors. A screwable framing material from wood-waste-reinforced inorganic
matrix was also developed. The resulting core and framing materials both
have economic advantages and offer a more reliable 90-minute fire door
than do those commercially available.
The final activity in Phase III was the construction of an 8-foot
wide, 12-foot long demonstration unit incorporating all of the above
products.
In Phase IV, wall panels constructed with the rice hull/inorganic matrix
composite were subjected to further structural tests to generate additional
data in support of possible future qualification for building code approval.
Structural tests consisted of racking shear, tranverse load, and long-
term weathering exposure.
Both a 1-hour fire test and a hose stream test were performed on a wall
assembly to complete the structural fire rating qualification initiated during
Phase III.
The material processing development work on fire door components was
continued. Detailed results of this work are presented in the following
sections of this report.
164
-------
SECTION 2
STRUCTURAL TESTS OF WALL PANELS
A structural test program on wall panels made of rice hull/inorganic
matrix composite was started during Phase III in the generation of data
required to qualify the wall panels for building code approval.
The requirements set by the International Conference of Building
Officials (ICBO) were used as the basis for the test program. ICBO is the
regulating body of the Uniform Building Code, one of the three major codes
used in the United States. The ICBO requirements were selected because
of MSC's prior experience with the ICBO qualification procedure (gained
during approval of its Duftec* wall panel system).
Four types of test are required for structural qualification of wall
panels under the ICBO requirements: axial compression, tranverse loading,
racking shear and impact test.
Considerable testing was conducted in Phase III,as reported in the
Phase III Summary Report; however, additional testing was considered
necessary to support qualification of the wall panels in order to obtain
approval of the maximum potential allowable design loads.
The additional tests conducted during Phase IV were transverse loading,
racking shear and long-term weathering exposure.
TRANSVERSE LOAD TESTS
Three 48- by 96-in. wall panels were tested in transverse loading in
accordance with ASTM-E-72 test methods. The tests were conducted by MSC
in its own test facilities under the supervision of San Diego Testing
Engineers Inc.
Results of the tests are presented in Figure 1. As indicated by these
load vs deflection curves, the wall panels are sufficiently stiff that
ultimate load, rather than the deflection criterion, determines the
allowable design loads. (The deflection criterion of L/180 would permit
a maximum deflection of 0.53 in., but all three panels failed at less
than half that value.)
*Duftec is a proprietary reinforced plastic material
developed by MSC for use in structual building components.
165
-------
40
30 .
Ul
r-t
I
Hi
W
$
H
20
10
.05 .10
DEFLECTION-inche s
.15
.20
.25
Figure 1. Transverse Load vs Deflection Curves for
Rice Hull/Inorganic Matrix Wall Panels
The average failure load of 32.67 Ib/ft will result in an allowable
transverse load of 10.89 Ib/ft based on a safety factor of 3. This is not
sufficient to satisfy the minimum wind load requirements of 15 Ib/ft speci-
fied by the Uniform Building Code.
In order to improve the transverse load (bending) resistance of the walls,
additional reinforcement would be required near the surface to increase
tensile strength of the outer layers in bending. Reinforcing fibers or fabrics
could be incorporated in a surface coating. The need for a surface coating,
which would provide moisture protection, will be discussed in the section on
the long-term weather exposure tests.
166
-------
RACKING SHEAR TESTS
Racking shear tests conducted in Phase III were made on v»ii
the top and bottom wood plates bonded with adhesive it wa^
these tests with the wood plates nailed ratneftnln ' bonded
Three 48- by 96-in. wall panels were tested in accordance with ASTM-P 77
test methods. The tests were conducted by MSC in its own test facilitLs
under the supervision of San Diego Testing Engineers Inc.
Results of the test are shown in Figure 2. From the corrected load-
vs-def lection curves, the allowable racking shear load is determined based
on the deflection or ultimate load criterion, as specified by ICBO
4000
3000
2000
1000
0.125 inches
I DEFLECTION LIMIT
0.1 0.2 0.3
CORRECTED NET DEFLECTION-inches
0.4
0.5
Figure 2 Racking Shear Load vs Deflection Curves for
Rice Hull/Inorganic Matrix Wall Panels
Nailed to Top and Bottom Plates
167
-------
Ultimate Load Criterion
The ultimate load obtained for each panel is divided by the width of
the panel and by a safety factor 3. The average of the three resulting
loads is the allowable load.
Panel #1: 3250 Ib = 270.83 Ib/ft
4 ft x 3
Panel #2: 3250 Ib = 270.83 Ib/ft
4 ft x 3
Panel #3: 3000 Ib = 250.00 Ib/ft
4 ft x 3
Average: Allowable Load = 263.89 Ib/ft
Deflection Criterion
The allowable design load based on the deflection criterion is the
average of the loads obtained at 0.125 inch deflection.
Panel ttl: 800 lb/4 ft = 200 Ib/ft
Panel #2: 800 lb/4 ft = 200 Ib/ft
Panel #3: 920 lb/4 ft = 230 Ib/ft
Average: Allowable load = 210 Ib/ft
Since the allowable load based on the deflection criterion results
in the lower value, it will govern design and would be the value approved
by ICBO.
LONG-TERM WEATHERING EXPOSURE TESTS
The results of structural tests conducted in Phase III indicated that
wall panels and other structural components developed in this program have
sufficient initial strength to satisfy design load requirements for one-
and two-story residential construction. The principal remaining question
was the effect of long-term weathering on the strength of these components.
Tests were devised, exposing typical wall panels to the following conditions.
(a) continuous axial compression load of 250 Ib/ft, normal outdoor
ambient exposure (Escondido, Calif.)
(b) continuous axial compression load of 250 Ib/ft, continuous rain
soak exposure (simulated by continuous water film running over one
exterior surface)
168
-------
(c) no load; repeated 24-hour cyclic exposure of one face to:
6 hours at -20°F
6 hours warm up to 120 F
6- hours at +120 F, with 1 hour water spray
at beginning and at end
6 hours cool down to -20 F
Tests (a) and (b) were performed in specially constructed loading fixtures
at MSC's Escondido facility (illustrated in Figure 3), Test (c) was performed
in MSC's cyclic- environmental chamber (illustrated in Figure 4).
Figure 3 Long-Term Exposure Test Fixtures
169
-------
Visual observations of wall panels in tests (a) and (b) indicated severe
surface cracking of all specimens. In the case of the water-soaked panels,
the surface of the wall degraded to a point at which it could be easily
punctured by the push of a finger.
Figure 4 Cyclic Environmental Chamber
170
-------
The axial and lateral deflection of the continuously loaded specimens
were recorded. (Deflection history curves for two rain-soaked panels are
presented in Figure 5). Deflection was substantial in the first 30 days
but appeared to level off after that. This would indicate that most of the
water-caused degradation occurs during the first month. Of course, the
deflection could be caused by warping due to the asymmetrical water exposure;
if so, once the wet face was completely saturated, the warping would stabilize.
There was no measurable deflection of the panels exposed to the outdoor
environment in Escondido.
0.3
w
I
o
G
•H
I
12
O
H
E-<
U
0.2.
0.1.
AXIAL DEFLECTION
LATERAL DEFLECTION
10
20
30
40
50
60
EXPOSURE TIME - days
Figure 5 Axial and Lateral Deflection of Rice Hull/Inorganic Matrix
Wall Panels During Long-Term Rain Soak Exposure,
Under Constant Axial Load of 250 lb/ft
One rain-soaked wall panel was removed for testing after 30 days and one
after 60 days of exposure. The walls were tested in axial compression follow-
ing a 30-day dry-out period. Test results (presented in Figure 6) show
a drastic reduction in both stiffness and ultimate strength after 30 days of
exposure. The wall exposed for 2 months shows some further reduction in
stiffness, but the failure load was actually slightly higher than that of
the 30-day specimen.
The difference between the 30- and 60-day results is not considered
significant; it is apparent that most of the degradation takes place in the
first 30 days of exposure to rain.
171
-------
30000
25000 •
T3
20000 -
15000
10000
5000
Typical Unexposed Panel
Normal Outdoor- 5 Months
Cyclic Chamber - 4 Months
Rain Soak - 30 days
Rain Soak - 60 days
Figure 6
0.05
DEFLECTION-inches
Axial Compression Load vs Deflection Curves
for 4- by 8-ft Rice Hull/Inorganic Matrix
Wall Panels After Various Long-Term Exposures
OT10
The two wall panels in the cyclic environment chamber were exposed for
about 4 months to the repeated test cycle (C); the chamber then was shut
down and the panels left in the chamber for 30 days to dry out. Visual
inspection of the exposed face of the panels indicated that the surface
cracks were similar to those of the rain-soaked panels.
172
-------
One of these panels was tested in axial compression (test results are
shown in Figure 6). Interestingly, this wall retained more of its original
stiffness and strength than the rain-soaked walls, indicating that water has
the most severe damaging effect on the rice hull/inorganic matrix composite.
One of the wall panels under continuous load in the outdoor environment
for 5 months was also tested in axial compression. This wall failed at
26,400 Ib, which is only a 12% reduction from the typical initial strength of
unexposed panels. This panel was severely cracked on the surface in a fashion
similar to the rain-soaked walls, but the material remained hard. The deflec-
tion curve shown in Figure 6 indicated that this panel retained most of its
original stiffness.
In conclusion, the results of the long-term exposure tests clearly indicate
that the rice hull-inorganic matrix material of the walls tested is very
sensitive to water damage. If these materials are to be used as building
components, they must be protected from prolonged exposure to water. It is
therefore recommended that future work should explore the development of
surface coatings. The coatings should perform two basic functions:
a) provide moisture protection and thereby prevent the degradation
of the rice hull/inorganic matrix composite;
b) reinforce the exterior surface of the panels against stress cracking
and thereby increase the tensile strength of the facings, which
will in turn improve the lateral load resistance (flexural strength)
of the wall panels.
173
-------
SECTION 3
FIRE TEST OF WALL
Several full-scale, ASTM-E-19 fire tests were conducted on wall assemblies
in Phase III of this program to obtain a 2-hour structural fire rating.
The results fell short of the 2-hour goal. The tests were repeated
with a 1-hour objective.
The first of the two-part test was successfully performed during Phase
III. The second part, the so-called hose stream test, was completed in
Phase IV: it was performed by the University of California Fire Test
Laboratory at Richmond, California. During this test, the wall assembly, 8
feet high and 12 feet wide, is exposed to fire for half an hour in accordance
with the procedures specified in ASTM-E-13 test methods. At the end of the
30 minutes, the temperature in the furnace reaches over 1500 F. One surface
of the wall panel is directly exposed to this temperature. At the same time
the wall is under a constant axial compression load of 100 pound per linear
foot.
The wall assembly is then removed from the furnace and subjected immediate-
ly to high pressure water spray, simulating fire hose exposure in real fire,
for 1 minute. The water must not penetrate through the panel for the wall
to be acceptable for 1 hour fire rating. In addition, at the end of the
hose stream test, the axial compression load is increased to twice the required
design load. The wall must not fail under the increased load.
The rice hull/inorganic matrix wall assembly successfully passed the
entire test and thereby qualified for a 1-hour structural fire rating with
an allowable design load of 1000 Ib/ft.
174
-------
SECTION 4
FIRE DOOR DEVELOPMENT
The work performed in the previous phases of this study resulted in the
development of new core and framing materials for use in fire doors.
After extensive material screening and processing studies conducted on
laboratory-scale samples at the end of Phase IV, components have been
fabricated successfully for full-scale fire doors.
Preliminary fire tests indicate that these new materials perform better
than commercially available door components. It also appears that in
commercial-scale production,the new materials would be competitive in cost
with currently used materials.
Having demonstrated the technical feasibility and the cost-effectiveness
of using waste materials in a commercial product(in this case a 90-minute
fire-rated door), the emphasis in Phase IV was concentrated on getting a
manufacturer interested in using the new materials in actual production.
Calwood Door, a cabinet and door manufacturer, has shown considerable
interest. Calwood assisted during the earlier phases of the development
effort by providing performance requirements and realistic full-scale
fabrication parameters to guide MSC in laboratory studies.
After the successful preliminary fire test, Calwood agreed to evaluate
full-scale production processing. Sample components were fabricated by
MSC and incorporated into full-scale doors by Calwood. The manufacturer
identified several production problems with the use of the new materials,
and MSC developed solutions.
For example, Calwood experienced difficulty in bonding the framing
components to the wood facing sheets. Laboratory tests were conducted by
MSC to find suitable adhesives to eliminate this production problem.
Results of tests performed on the two most promising candidate adhesives
are presented in Table 1. The test consisted of bonding blocks of the
framing material to wood block using the candidate adhesives. The bonded
blocks were then tested in tension at room temperature and at 300 F. The
use of "veil mat" as a reinforcement of the bond interface was also evaluated
in these tests.
From the results of these tests, it was concluded that both of the adhesives
would perform satisfactorily. The results also indicated that the use of the
175
-------
veil mat did not increase the strength of the bond. In fact, at elevated
temperature,the specimens with vail mat showed lower strength than the ones
without it.
Based on these results, the manufacturer repeated his earlier production
experiments with the new adhesives and found that they both performed well.
Table 1 FIRE DOOR ADHESIVE TESTS
Borden 07-100 (Casein) Adhesive
Type of Failure
Block Broke
Without Veil Mat
Failure
Load, Ib @ R.T.
615
495
780
620
525
Avg 607
Tested @ 300°F
420
515
440
520
490 "
Avg 477
22% loss in material strength
With Veil Mat
Failure
Load, Ib @.R.T. Type of Failure
Block Broke
630
660
635
605
610
Avg 628
Tested @
380
Block Broke
300°F
Block Broke
420
455
430
Avg 436 "
31% loss in material strength
Borden RS-125 MD (Resorcinol) Adhesive
Without Veil Mat
Failure
Load, Ib @ R.T.
Type of Failure Load, Ib
With Veil Mat
Failure
@ R.T. Type of Failure
620
630
750
575
650
Avg 644
Tested @
535
430
520
515
470'
Avg 494
23% loss
Block Broke
95% Adhesive Failure
Block Broke
ii it
ii ii
^*
300 F
Block Broke
ii ii
H It
II II
It II
in material strength
565
685
645
705
575
Avg 635
Tested @
415
450
420
460
455
Avg 440
31% loss
Block Broke
it H
it n
ti n
n n
300°F
Block Broke
it n
n it
ii it
n n
in material strength
176
-------
As a conclusion to this study a door fabricated with waste material
components developed in this program was fully qualified as a 90-minute
rated fire door.
Based on the results of this study, Calwobd Door is planning to intro-
duce in next year's product line a new high-performance fire door using
components made from waste materials.
177
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/2-78-1
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
A STUDY OF THE FEASIBILITY OF UTILIZING SOLID WASTES
FOR BUILDING MATERIALS
Phase III and IV Summary Reports
5. REPORT DATE
May 1978 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Material Systems Corporation
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Material Systems Corporation
Escondido, California 92025
10. PROGRAM ELEMENT NO.
1DC618 SOS #5 Task 08
11. CONTRACT/GRANT NO.
68-03-2056 (Research)
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory--Cin.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final Sept. 1975 - March '78
14. SPONSORING AGENCY CODE
EPA/600/14
is.SUPPLEMENTARY NOTES Project Officer: Robert E. Landreth (513) 684-7871".
See also Phase I (EPA-600/2-78-091, PB-279 440); Phase II (EPA-600/2-78-092,
PB-279 441); and Executive Summary (EPA-600/8-77-006, PB-271 007).
16. ABSTRACT
This report summarizes work to develop building materials containing inorganic and
organic wastes and wastes-derived products. Attempts were made to produce full-scale
products and qualify them for structural applications. Particle board panels were
made of peanut hulls and wood waste on production-type equipment. Particle boards
of peanut hulls* have mechanical properties that are slightly less desirable than
those of commercially available boards, and the economics are marginal. However,
particle board panels of wood waste can be competitive with commercial products.
Two hour, fire-rated structural walls made from inorganic rice hull foam could also
be viable, as could floors, roofs, ceilings, and the 90-minute, fire-rated door with
a rice-hull foam core and a wood-waste frame. Data were developed for submittal to
the International Conference of Building Officials and the U.S. Department of Housing
for certification. Structural tests were performed on wall panels fabricated from
rice hulls and an inorganic binder. These tests completed generation of the data
required for building code approval. Wall panels made of rice hull composite were
found to satisfy all structural requirements necessary for use in one- and two-story
residential construction, except that resistance to transverse loading (bending) was
insufficient.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Group
Construction materials
Fire resistant materials
Materials recovery
Agricultural wastes
Solid wastes
Rice hulls
Resource recovery
13B
8. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
190
20. SECURITY CLASS (This page)
22. PRICE
UNCLASSIFIED
EPA Form 2220-1 (Rev. 4-77)
178
U. S. GOVERNMENT PRINTING OFFICE: 1978 —757-140/1363
------- |