<%BANNER%>

Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-12-31.

Permanent Link: http://ufdc.ufl.edu/UFE0041269/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-12-31.
Physical Description: Book
Language: english
Creator: Kelley, Kristy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Building Construction -- Dissertations, Academic -- UF
Genre: Building Construction thesis, M.S.B.C.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Statement of Responsibility: by Kristy Kelley.
Thesis: Thesis (M.S.B.C.)--University of Florida, 2009.
Local: Adviser: Muszynski, Larry C.
Local: Co-adviser: Issa, R. Raymond.
Electronic Access: INACCESSIBLE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041269:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041269/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-12-31.
Physical Description: Book
Language: english
Creator: Kelley, Kristy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Building Construction -- Dissertations, Academic -- UF
Genre: Building Construction thesis, M.S.B.C.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Statement of Responsibility: by Kristy Kelley.
Thesis: Thesis (M.S.B.C.)--University of Florida, 2009.
Local: Adviser: Muszynski, Larry C.
Local: Co-adviser: Issa, R. Raymond.
Electronic Access: INACCESSIBLE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041269:00001


This item has the following downloads:


Full Text

PAGE 1

1 USE OF RECYCLED OYSTER SHELLS AS AGGREGATE FOR PERVIOUS CONCRETE By KRISTY NOEL KELLEY A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN BUILDING CONSTRUCTION UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Kristy Noel Kelley

PAGE 3

3 To the power of perseverance

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank Dr. Muszynski for a lways believ ing in the potential of both a stinky shell and a disabled wrist. I would like to thank Dr. Issa for always letting me keep coming back for another semester. And I woul d like to extend my appreciation to Dr. Stroh for being patient and sticking with me in this thesis even though it changed routes. Finally, I thank my village those who have raised this graduate student; be you friends near or far, boyfriend or family, I could not have made this journey without you. Thank you for reminding me to never, never give up.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDG MENTS .................................................................................................. 4LIST OF TABLES ............................................................................................................ 7LIST OF FIGURES .......................................................................................................... 8LIST OF ABBR EVIATIONS ............................................................................................. 9ABSTRACT ................................................................................................................... 10 CHA PTER 1 INTRODUC TION .................................................................................................... 12Research Overview................................................................................................. 12Goals and Obje ctives .............................................................................................. 162 BACKGRO UND ...................................................................................................... 17Oyster Shells in Constr uction .................................................................................. 17Quicklime .......................................................................................................... 17Tabby ............................................................................................................... 17Shellcre te ......................................................................................................... 23Roads ..................................................................................................................... 24Current Uses ........................................................................................................... 25Pervious C oncrete .................................................................................................. 25History .............................................................................................................. 25Composit ion ..................................................................................................... 26Applicat ions ...................................................................................................... 27LEED ................................................................................................................ 29Standards ......................................................................................................... 303 METHODOLOGY ................................................................................................... 31Materi als ................................................................................................................. 31Oysters ............................................................................................................. 31Cement ............................................................................................................. 31Water ................................................................................................................ 31Oyster Shell Acquisitio n, Cleaning and Crus hing .................................................... 31Acquisiti on ........................................................................................................ 31Cleanin g ........................................................................................................... 32Crushing ........................................................................................................... 33Pervious Concrete Mix Design ................................................................................ 36

PAGE 6

6 Introduction ....................................................................................................... 36Sieve An alysis .................................................................................................. 36Unit We ight ....................................................................................................... 37Specific Gravity ................................................................................................ 37The Ball Test .................................................................................................... 38Mixing and Compacti on Tec hniques ....................................................................... 39Introduction ....................................................................................................... 39Mixing ............................................................................................................... 40Compacti on ...................................................................................................... 40Rodding ............................................................................................................ 42Vibrat ion ........................................................................................................... 42The Proctor Method .......................................................................................... 43Pervious Concrete Testing ...................................................................................... 44Unit Weight for Fresh Conc rete ........................................................................ 44Water Displacem ent Test ................................................................................. 45Permeability Constant Head .......................................................................... 46Permeability Falli ng Head Appa ratus ............................................................. 47Sulfur Capping .................................................................................................. 48Compression Testing ....................................................................................... 494 DATA ANALYSIS AND RESU LTS .......................................................................... 50Aggregate A nalysis ................................................................................................. 50Identify a Viable Conc rete Mix Design .................................................................... 51Early Tr ials ....................................................................................................... 51The Ball Test .................................................................................................... 51Pervious Concrete Testing ...................................................................................... 52Void Content ..................................................................................................... 52Compre ssion .................................................................................................... 53Permeability ...................................................................................................... 555 SUMMARY AND CO NCLUSION ............................................................................ 58Complicat ions ......................................................................................................... 58Conclusi ons ............................................................................................................ 58 APPENDIX RESULTS OF OYSTER SHELL SIEVE ANALYSIS ............................... 60LIST OF RE FERENCES ............................................................................................... 61BIOGRAPHICAL SKETCH ............................................................................................ 64

PAGE 7

7 LIST OF TABLES Table page 2-1 Characteristics of orig inal tabby and new tabby ................................................. 233-1 Pervious oyster shell c oncrete. ........................................................................... 384-1 Properties of the Eastern Oyster Shell. .............................................................. 514-2 Specific Gravity and porosity by compaction method on samples placed in 4x8 molds. ........................................................................................................ 534-3 Compressive strengths and permeability rates of one-year samples in 4x4.5 molds. ................................................................................................................. 564-4 Compressive strengths and permeabilit y rates of 7-day and 21-day samples placed in 6x6 molds. ........................................................................................ 574-5 Compressive strengths and permeabilit y rates of 7-day and 1-year samples placed in 4x8 molds. ........................................................................................ 57

PAGE 8

8 LIST OF FIGURES Figure page 1-1 U.S. raw nonfuel mineral materials put into use annua lly, 1900-2006. ............... 132-1 Barn of tabby construction on King sley Plantation, St. George Island, Jacksonville FL. ................................................................................................. 192-2 The Ashantilly House, Sapelo Isla nd, GA. .......................................................... 202-3 Tabby concrete shown through cracks in the st ucco faade. ............................. 222-4 Centennial House, Co rpus Christ i, TX. ............................................................... 242-5 Pervious concrete. .............................................................................................. 272-6 Pervious pavement parki ng lot. .......................................................................... 283-1 Shell cleaning. .................................................................................................... 333-2 Crushing shells at the Florida De partment of Transportation Materials Research Lab. .................................................................................................... 343-3 Using a Caterpillar Skid Steer for oyster s hell crus hing ...................................... 353-4 Early moldi ng techniques. .................................................................................. 423-5 Vibration of a 6x6 sa mple. ................................................................................ 433-6 The customized Standard Pr octor compacti on method. ..................................... 443-7 Permeability testing ............................................................................................ 463-8 Concrete specimen inside the falling head permeab ility apparat us. ................... 483-9 Sulfur caps being applied to the pervi ous oyster shell concrete samples. .......... 483-10 Compressi on test ing ........................................................................................... 494-1 Relationship between porosity and compaction. ................................................. 534-2 Relationship between compressi ve strength and compaction. ........................... 554-3 Relationship between permeability and co mpaction. .......................................... 56A-1 Results of test usi ng sieve sizes 1#30. ......................................................... 60

PAGE 9

9 LIST OF ABBREVIATIONS ASTM American Society for Testing and Materials ACI American Concrete Institute W/C Water to cement ratio A/C Aggregate to cement ratio PCC Portland Cement Concrete AC Asphalt Concrete

PAGE 10

10 Abstract of Thesis Pres ented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for the Mast er of Science in Building Construction USE OF RECYCLED OYSTER SHELLS AS AGGREGATE FOR PERVIOUS CONCRETE By Kristy Noel Kelley December 2009 Chair: Larry Muszynski Co-Chair: R. Raymond Issa Major: Building Construction Increasing costs of transporting ra w aggregate and the high demand of the construction industry on the Earths natural resources have led researchers to begin investigating alternate sources for concre te aggregates. Recycled content aggregate is one area that has gained popular ity in that arena. In this study, recycled oyster shells were used as an aggregate for pervious concrete. The shells used in this study we re diverted from a local restaurants waste stream. It was found t hat the restaurant sent approximately 10,000 oyster shells to the landfill each week. The shells were cl eaned and crushed and a pervious concrete mix design was developed. Six compactive methods were used to place the concrete including: vibrated, rodded, Standard Proctor, customized Standard Proctor and Modified Proctor, as we ll as non-compacted. The results showed a potential for a viable pervious oyster shell concrete with appropriate compaction methods. Porosity ranged from 26-38%. Compressive strengths had a very wide range of 206 ps i with a vibration-compact ed sample to 1596 psi with a Modified Proctor sample. Permeability also had a wide range, with drainage rates of 27

PAGE 11

11 gal/min/ft with a vibration-co mpacted sample, to 0.3 gal/min/ft with a Modified Proctor sample.

PAGE 12

12 CHAPTER 1 INTRODUCTION Research Overview It has been estimated that Americans will use as much aggregate over the next 25 years as they did cumulatively over the whole of the 20th century (Langer et al. 2004). In 2007, the US led the world in nonfuel mineral mining and processing with values totaling $575 billion (Hitzman et al ., 2009) The construction industry alone consumes about 70% of the overall mineral ex traction in the United States in order to build and maintain commercial and residentia l structures, highways, bridges and parking lots. The United States Geological Survey, in a 1998 report Aggregates from Natural and Recycled Sources defines the term aggregat e as, materials, ei ther natural or manufactured, that are either crushed and combined with a binding agent to form bituminous or cement concrete, or treat ed alone to form products such as railroad ballast, filter beds, or fluxed material. Approximately three billion metric tons of aggregate are used in construction per year (Shulman, 2005). While this number may sound surprising, it may be rationalized when the following numbers ar e taken into account: It takes 38,000 tons of aggregate to ma ke one mile of a four-lane highway. The American average single-family home requires 400 tons of aggregates to build. A 100,000 square foot office building r equires 5,000 tons of aggregate (Wilson, 2008).

PAGE 13

13 Figure 1-1. U.S. raw nonfuel mineral materials put into use annually, 1900-2006. Source: Used with permission by Matos, G. (2009). Use of Minerals and Materials in the United States From 1900 Through 2006, Fact Sheet 20093008. U.S. Geological Survey, Reston, VA. In one study by Purdue University, data we re compiled showing that extraction of natural aggregates for the concrete and cem ent industry as being responsible for only 18% of the total waste and mi ning overburdens in 1996. Ye t that small percentage was still a staggering 447 MMT (million metr ic tons). It was also estimated that the ratio is 6:1 for the ratio of total raw material moved to a given volume of concrete (Low, 2005). Aggregates are generally mined or qua rried from more than 9,000 pits and quarries across the country. (Shulman, 2005) Though it may stand to reason that the Earth is made of rock therefore aggregates should be a relatively easy material to acquire, this is not always the case. Some areas may have a high water table and make extraction expensive. Other areas may have a good deposit of limestone, for example,

PAGE 14

14 but its source is at such a depth as to make it impractical. Some aggregates simply may not react well chemically when used in appl ications such as concrete or asphalt. Transporting aggregates generally has a high cost impact, so being close to urban centers has significant economic benefits. There ar e, however, many environmental concerns associated with establishing a mine or quarry near a densely populated locale. Some of these concerns include: Increased dust, noise and blast vibrations High density of tr ansport truck traffic Physical landscape changes and wil dlife habitat disturbances Possible pollutants added to groundwater and soil Emission of greenhouse gases into the air There are alternatives to using only raw material aggregates in construction. The use of recycled aggregate is one way to potentially extend the life of natural resources by supplementing their supply, reduce the env ironmental impact of material extraction, as well as the impact of construction demo lition in landfills. According to a U.S. Environmental Protection Agency study, construction and demolition waste totaled more than 135 MMT in 1998. Of that amount, as much as 75% could have been diverted to recycling centers (Kibert, 2005). Recycled aggregates are obtained by crushing concrete and often asphalt. Recycled aggregate comes primarily from Port land cement concrete (PCC) and asphalt concrete (AC). The concrete in its or iginal state generally comes from road rehabilitation, maintenance or demolition, structure demolition, and leftover batches of AC and PCC. Recycling of constructi on aggregate is most commo n near urban areas where constant replacement of infrastructure is necessary, natural resources are limited or

PAGE 15

15 uneconomical to bring in, demolition disposal co sts are high or regulations, be they local or federally mandated environmen tal policies, prevent their disposal (Wilburn et al., 1998). Aggregate recycling centers are increas ing in number, but the demand is great and the supply slow. Approximately only 5% of the 3 billion metric tons (BMT) of new construction aggregate used in 2005 was comp rised of recycled ag gregate (Shulman, 2005). This number can be expected to ri se with the public demand for more sustainable materials becoming an ever-incr easing factor in materials acquisition, concrete and asphalt mix designs becoming lighter, stronger and cheaper with the inclusion of recycled content, and the introduction of portable, on-site recycling centers. Other options in the recycled aggregate market aside from post-construction concrete or asphalt are constantly being in troduced. Post-consumer glass, fiberglass pellets, plastics, even old tires are ma king a more pronounced presence in the aggregate arena. However, t here are obstacles for these recycled aggregates to overcome as well. For example, glass must be sorted according to color, broken into smaller pieces then crushed, sorted and cleaned or in some cases, melted. Fortunately, much of the construction industrys uses for glass aggregate do not require the laborand time-intensive sorting process. In 1993, New York City collected 27,000 tons of mixed waste glass and used it to produce glasphalta 90% asphalt and 10% glass composite mixture (Michigan Technological University, 2003). The majority of the alternative recycled aggregates are derived as a postconsumer resource, which means that public par ticipation is vital. This is perhaps the biggest setback to alternate aggregates. In 2007, Americans generated about 254

PAGE 16

16 million tons of trash, of which only 85 million tons were recycled or composted (EPA, 2007). This is the equivalent of a 33.4% recy cling rate an 8% increase from the 26% reported in 2002. It is expected that this rate could climb to as high as 60-80% with energy prices for production continuing to rise and national resources facing depletion (Chiras, 2002). Conclusion The success of recycled construction aggregate and alternative aggregates depend largely on the involvement of the indivi dual; without the consumer, there would be no post-consumer content. And the consumer is not just the average citizen disposing of their trash and deciding to sort it into recycling bins; it is also the building owner who demands materials used on their jobsite that inco rporate a certain percentage of recycled content; and it is the aggregate company that begins to offer more recycled content options. It is also about the government inst ituting policies that help ease the transition as non-conventional aggregates take a more active role in the construction industry. Goals and Objectives This study evaluates the potential of the Eastern Oyster shell as a recycled aggregate, as well as seeks to develop a pervious oyster shell concrete mixture that is a viable, sustainable building materi al. It is the intent of this research to create as little environmental impact as possible during the course of this study. It is also the intent of this study to create an uncomplicated mix desi gn, which may be of particular use to individuals and small-scale projects.

PAGE 17

17 CHAPTER 2 BACKGROUND Oyster Shells in Construction Quicklime Perhaps one of the oyster shells mo st popular uses in construction throughout history has been in its burnt fo rm as lime, also known as quicklime. Documentation of oyster-based lime dates back several thousand years and crosses many nations and continents including Honduras, Brazil, Gr eece, Northeast India, and Australia. One early American reference to lime burning is found in Mechanick Exercises by Joseph Moxon, published in 1703. Moxo n detailed his theories on the various trades including carpentry, and bricklaying. In his discussion on lime, he states, But the shells of Fish, as of Cockles and Oysters are good to burn for Lime (Sickels-Taves, 1996). To make quicklime, a pit was excavated and filled with wood, preferably pine or heart pine. Once the pit was completed a small log structure, also known as a rick, was built on top. This rick contained several tier s of logs within which were held smaller layers of logs covered with oyster she lls. The entire structure was set ablaze and allowed to collapse upon itself. Interior temperatures reached approximately 2000F. These high temperatures were necessary to ensure the proper chemical reactions necessary to convert the calcium carbonate of the shell into the calcium oxide of lime (Sheehan et al., 1998). Tabby Lime obtained from oyster shells first appeared in mortar mixes during the Middle Ages, apparently originating in North Afri ca and Spain. This mix (and similar mixes using on-hand aggregates) wa s given several names: pise in French influenced areas,

PAGE 18

18 tapia (meaning mud wall), in the Spanish-speaking world, tabya in Islamic countries and tabby in the British world. The latter is t he most commonly used name, and its origin could be an adaptation of tapia or the Arabic word tabbi which means a mixture of mortar and lime. There are also similar words appearing in both Portugese and Gullah. (Morris, 2005) Tapia was said to be one of the most common building ma terials of the 15th century Cordoba and Seville and a standard construction method in many Muslim territories in the 13th century, especially for m ilitary purposes. (Deagan, K. 2002) There are even several examples of buildin gs made out of oyster shells using a tabby material in China. These homes are called Erkecuo houses and can be dated back to the Qing dynasty (1644-1911). It is believed that the oysters used in the Erkecuo homes were from Afri ca because the shells are appr oximately 3-5 times larger than local shells. Historians propose the s hells were used as ballast on trade ships returning back to the port of Xunpu. The Xunpu village was on the Maritime Silk Road, trading silk, tea and pottery (WOX Info, 2009). The most common examples, however of tabby-like structures are from Europe. One example of British tabby building is t he Wareham Castle, in Dorset, England dating from the early 11th century. The building is in ruins, but was excavated in the 1950s by H.J.S. Clark (Renn, D.F., 2009). The castle is now a unique structure for the British, as disease has all but destroyed their oyster industry. When the Spanish and English began to build up the American colonies they carried the art of oyster conc rete with them. There were two primary centers for its

PAGE 19

19 distribution: British-built t abby arising out of Beaufort, South Carolina while Spanish traditions were derived from St. Augustine, Florida (Adams, D., 2007) Figure 2-1. Barn of tabby construction on Kingsley Plantation, St. George Island, Jacksonville, FL. James Oglethorpe, a British officer stationed in South Carolina in the early 1700s is credited with bringing tabby into Geor gia. Oglethorpe believed it to be a logical material with which to build fortifications. Therefore, with his backing forts, support structures, and even his own home were built out of tabby. However, with the Treaty of Aix-la-Chapelle in 1748, t he threat of Spanish invasion ceased, and the use of "Oglethorpe tabby" diminished. Approximately 90 years later, another advocate of tabby would arise by the name of Thomas Spalding. Spaldi ngs father had purchased Oglethorpes Tabby House and Thomas had been born there. Tabby from that era is oft en called Spalding Tabby. Spalding built a home and m any other structures on S apelo Island, Ga, where he was a prominent businessman. His house is also referred to as the Tabby House or the Ashantilly house. Spalding kept copious notes and journals regarding the building methods, seed cultivations and farming techniques employed at the ti me. In addition, he

PAGE 20

20 promoted the use of Tabby in local publicat ions and other various periodicals. These writings provide some of the most detailed information available regarding the history of Tabby. In one article, Spalding wrote, Tabby is a mixture of shells, lime and sand in equal proportions by measure and not we ight and makes the best and cheapest buildings, where the materials are at hand, that I have ever s een; and when rough cast equals in beauty stone, (Sullivan, B., 1998). Figure 2-2. The Ashantilly House, Sapelo Island, GA. The spread of tabby th roughout the south between t he early 1800s and the mid1800s was largely tied into the creation and expa nsion of plantations. With the onset of the civil war, the use of t abby waned. After the war, the traditional method for making tabby was altered because of the introduction of Portland cement to the United States. The use of Portland cement negated the need for lime and was preferred because it completely removed the arduous task of burning oyster shells from the tabby process. Tabby made with Portland cement is most co mmonly referred to as Tabby Revival and was used, though in much smaller am ounts, between approx imately 1870-1925 (Sickles-Taves, 1999).

PAGE 21

21 By the end of the 19th cent ury, the use of tabby as a building material was almost nonexistent. There are several factors that appear to have been significant in its decline: first, construction of new buildings severely declined during Civil War years; second, the breakdown of t he plantation system denied owner s the plethora of unskilled workers necessary for the labor-intensive process; and third, the introduction and easy availability of Portland cement and readymade concrete blocks. Traditional tabby is a slurry of equal parts lime, sand, water and shell. These materials are measured by volume, not wei ght. Thomas Spalding wrote his formula as bushels of lime, 10 bushels of sand, 10 bushels of shells and 10 bushels of water to yield 16 cubic feet of wall. He made wa lls on average 14 inches thick. Those beneath the ground were 24 inches and the second st ory walls were 10 inches; he would not erect tabby buildings beyond two stories (Adams, D., 2007). The tabby slurry was poured into wooden cr adle forms that were held together by round pins. Early tabby forms were generally 20-22 inches in height but by the 19th century were reduced to 10-12 inches to reduce chance of collapse and provide greater strength (Sickles-Taves, 2008). Once in t he form, the slurry was hand tamped and leveled. The tabby was then air dried in it s cradle for several days before the next layer was added on top. Although generally poured into wall fo rms, tabby can be formed into many various architectural features. By crushing t he oyster shells, smaller wedges and bricks could be made, with a tabby mortar employ ing finely crushed shells used to cement them together.

PAGE 22

22 Tabby is a very porous material and meant to be covered by stucco for protection. Traditionally, the stucco mix is just a thicke r mixture of the tabby but without the oyster shells, however, stucco containing crushed shells have been documented. Figure 2-3. Tabby concrete shown through cracks in the stucco faade. Cracks in the stucco faade resulting from settling over time or weather damage such as hurricanes, improper application of the stucco or lack of maintenance can all be causes of stuccos failure to prevent wate r intrusion into the tabby material. Trenching around the foundation of a struct ure and filling it with oyster shells was one way to further the life of a stucco coat ing as the shells would wick away moisture from the base of the walls. It is a paradox for tabby to be a very porous material yet be native to salty coastal areas. Fortunately, the tabby can generally control the absorption and evaporation with the help of the warm climate. This process is significantly slowed or even brought to a halt if incompatible materials are introduced onto the tabby that makes it impossible for this cycle to continue. Such was the case with the Oglethorpe Tabby House on Cumberland Island, Ga. In the early 1990s, neat cement had been added as a stucco layer to the surface of the

PAGE 23

23 tabby. It was soon discovered that the tabby couldnt breathe. The walls began to buckle, mold and mildew appeared and paint began to peel on interior walls. The National Park Service was forced to install a 24-hour fan to circulate air until the repairs could be mitigated. Dr. Lauren Sickels-Taves, an expert on t he care and preservation of tabby, did numerous studies on the Tabby house. The thr ee of the tests she ran were the particleinduced X-ray emission (P IXE) test, an optical stereology and an ash analysis. The PIXE test provided a chemical analysis of the tabby, the optical stereology identified the sand used in the mix as deposits from a lo cal channel, and the ash analysis confirmed the inclusion of wood ash from th e lime rick (Sickles-Taves, 2008). During testing, Sickels-Taves also took core samples from some of the original tabby and compared it to a sample of tabby m ade in the laboratory. The results were as follows: Table 2-1. Characteristics of original tabby and new tabby Original Tabby New Tabby Compressive Strength (psi) 350 320.5 Specific Gravity 2.013 2.203 Using these tests, Sickels-Taves identif ied the formula for the Oglethorpe Tabby House as a 1:3 lime: sand with wood ash, obtained during the process of burning the oyster shells. This specific formula is perhaps ironically, one of two formulas for concrete specified by the ancient Roman Marc us Vitruvius Pollio in his famous work dated 4630 BC, The Ten Books of Architecture (Sickels-Taves, 1998). Shellcrete Another historic oyster concrete is called shellcrete, indigenous to Spanish Texas and popular in the mid-1800s. Shellcrete was made by creating bri cks out of a mixture

PAGE 24

24 of crushed shells, lime, water and thick, wh ite local clay. A small amount of sand was used to temper the clay if necessary. The mixture was tamped into sand-dusted forms, sun-hardened then baked in a firewood-heated kiln (Givens, 2006). One example is the Centennial House, al so known as the Forbes Britton House, in downtown Corpus Christi. Built in 1849, the homes stout 3-layer brick construction has reportedly provided shelter to the to wns inhabitants through many hurricanes and even Mexican bandit raids (Cox, 2006). Figure 2-4. Centennial Hous e, Corpus Christi, TX. Roads As the demand for oyster shells was decr easing in the home bu ilding industry, it was increasing elsewhere in t he world of construction. O ne 1894 U.S. Department of Agriculture public road bulletin praised the s hells value and its cementitious capacity. The author sited large midden piles as a near-endless resource for aggregate (Stone, 1894). However, when those supplies r an out, Florida began dredging coastal waterways for shells. Those oyster shells were pulverized then used as a main ingredient in highway paving for many years in Florida. This practice was in use until the

PAGE 25

25 mid-80s, when legislation called for its halt d ue to the detrimental nature of dredging (Berrigan, 2009). Current Uses In Florida at this time, 50% of a ll commercially harvested shells, with the exception of those sent to hal f-shell restaurants, must be returned to the state for its reef reconstruction program (Florida Statues Title XXXV, 2009). The remainder of all shells may be sold to the highest bidder, which is most commonly the poultry feed industry. Many coastal residents choose to use oyst er shells as pavement for driveways. The shells are often just placed on the ground and not mixed with any cementitious material. Also, some enterprising individuals have begun creating artificial fish tank structures out of crushed s hells. And there are even some who go the extra mile and use a modern tabby formula to recreate the old style as closely as possible while successfully maintaining modern code standards. Pervious Concrete History In addition to the search for new or mo re sustainable aggregates for concrete is also the quest for a more environmentally friendly pavement. One option that is becoming more popular called pervious concrete A relatively new material in the US, pervious concrete first appeared in the United Kingdom in 1852 as a structural material for building homes, then later as a pavement in the mid-1960s. In the US, it has acted as an environmentally friendly alternative in stormwat er management practices, a champion for fertile, tree-lined parking lots and a friend to those in need of a sturdy, non-slip surface for more than 20 years.

PAGE 26

26 Pervious concrete, also known as no-fines or gap-graded concrete was a very popular building material in Europe duri ng the post-World War II construction boom. Both skilled labor and building materials were in short supply so an English man named George Wimpey developed a method that requi red men with little or no skill to erect forms and place no-fines concrete to build 2-story homes (Offe nberg, M., 2008). It is unknown when pervious concrete was first introduced into the United States, though the earliest documented pervious concrete pavement at this time dates back to 1985. The National Ready Mix Concrete Association (NRMCA) is currently compiling a record of all U.S. pervious projects. T he database now lists approximately 250 entries with more than 60% in the Southeastern stat es and 20% in California. The rest are spread across the country and include states such as Illinois, Pennsylvania, Ohio and Vermont (Kresge, 2009). Composition Pervious c oncrete is a composite mate rial containing Portland cement, water, coarse aggregates and little or no fine aggregates that allows water to pass through it. There is generally no slump and a very low water-to-cement ratio, as well as a low aggregate-to-cement ratio; usually just enough to provide adhesion of the aggregate but not enough to lose porosity. These low ratios contribute in part to pervious concrete being both a lightweig ht and sustainable product, as fe wer materials are necessary for its manufacture. Being held together at contact points by the thin cement paste is what gives pervious concrete its voids. Average void contents range from 20-30% depending on what type of aggregates or compaction techni ques are used (Kosmatka et al., 2002). Voids and compaction methods also determine the permeability of pervious concrete.

PAGE 27

27 There is a wide range of acceptable drainage rates, which are based for the most part on its intended application. An ACI R-06 report on pervious concrete states that drainage rates may vary between 2-18 gal/min/ft Compressive strengths in pervious concrete are also quite variable a nd can range from 400 psi to 4000 psi. Figure 2-5. Pervious concrete. Applications Pervious c oncrete can be used in a wide range of applications throughout the construction industry. Perhaps t he most popular use in the U. S. is as pavement for lighttraffic volume parking lots and sidewalks, though it has become more prevalent in stormwater systems. Pervious concrete has also been developed that is highly effective for use as flooring in greenhouses, as well as artificial reef structur es (ACI R-06, 2006). Use of pervious concrete as a pavement is considered an environmentally sound approach to the challenges that face many builders and owners when faced with stiff local and federal stormwater run-off regulations. Currently the EPA requires owners of sites of 1 acre or more development to have an on-site stormw ater management system in place. This often leads to 10-20% of t he gross buildable area dedicated to non profitgenerating features such as holding ponds or swales (Huffman, 2005). Because the

PAGE 28

28 EPA considers pervious concrete a Bes t Management Practice, those large percentages can be significantly reduced by combining the parking lot with the stormwater treatment system. Figure 2-6. Pervious pavement parking lot. Pervious concrete is dubbed a stormwater treatment system in part because of the way it virtually eliminates the first-flush action of normal pavements. After heavy rains, the first flush sends surface toxins and pollutants along the pavements horizontal surface and into the nearest drainage outlet, then eventually streams and lakes. With a pervious concrete system, rain water does not pool or pond, nor does it allow water to rapid flow across its surface. Instead, the water and surf ace toxins drain through the concrete, then into the sub base and eventually into the soil below where further filtration may take place. This process provides a water purification system for stormwater runoff, as well as a groundwater recharge. As it was stated earlier, one of the most popular applications for pervious concrete is as a pavement. There are many benefits to using pervious concrete as a pavement instead of conventional asphalt, including:

PAGE 29

29 Reduced glare means better visibility No puddles or standing water l eads to less dangerous hydroplaning Larger air voids prevent pumping acti on, thereby reducing tire temperature Better sound insulation means less road noise Popcorn-like texture gives ti res a better grip on the road LEED Pervious c oncretes green characteristi cs often help earn buildings high ratings in several LEED categories. Leadership in Envi ronmental Energy and Design (LEED) is the green-building standard in the United States and Canada that was developed to give professionals a template from which to identify and implem ent sustainable building planning, design, construction, and maintena nce practices. Potential LEED points for pervious concrete use may be available in the areas of stormwater management, landscape paving, minimizing energy use, re cycled content, use of regional materials, site-wide VOC reduction, reduction in the us e of Portland cement, and innovation in design. To qualify for category MR 4.1 & 4.2, it is requi red that the sum of post-consumer recycled content plus one-half of the pre-consum er recycled content constitute at least 10% or 20% respectively, (based on the dollar value of the material ) of the total value of materials in the project. The requirement for post-consumer content is easily met with oyster shells, as they are directly divert ed from the public waste stream. However, if following the mix design as tested for this research, the pre-c onsumer requirement would not be mixed. This could be resolved during future research with the addition of supplementary cementitious materials such as fly ash or blast furnace slag, which are pre-consumer recycled material.

PAGE 30

30 Standards With the rise in popularity of pervi ous concrete has come the need for quality control. Currently the ASTM Subcommitt ee C09.49 on Pervious Concrete has been reviewing a wide range of test methods devel oped all across the worl d. Members of the committee have been reviewing procedures fo r compressive strengths, flexural strengths, in-place permeability and in-p lace density/porosity. The Subcommittee released ASTM C1688 in late 2008, which provides a standard method for testing the unit weight of fresh pervious concrete.

PAGE 31

31 CHAPTER 3 METHODOLOGY Materials Oysters The aggregate used in this project is shell of the Eastern Oyster ( crassostrea virginica ), which is found in the Am erican Atlantic and Gulf waters. This species of oyster is the most commonl y found variety in the Southeastern United States and is predominately the only one sold at half-shell restaurants. Cement The cement chosen for this project was Type I Portland cement, which is perhaps the most common type of cement available on the market today. This product was chosen primarily for its easy accessibility, economic appeal, as well as its general durability. Water Only potable water adhering to ASTM C1602 was used in this research. Oyster Shell Acquisition, Cleaning and Crushing Acquisition Several methods were attempted in the effo rt to obtain a large quantity of shells. Fliers were distributed to various University of Florida buildings as well as nearby businesses, requesting anyone with waste shells to deliver them to a designated dropoff or call for pick up. No shells were obtained through this method. Several Gainesville seafood wholesalers were approached only to find that they are not legally allowed to shuck their own oysters.

PAGE 32

32 It became clear that the remaining method of obtaining oyster shells was to rely on a half-shell restaurant. In Gainesville, FL, which in not a coastal community, there is only one restaurant of consi derable size and that is Calico Jacks (CJs). Upon communication with the restaur ants management, it was learned that approximately 10,000 are shucked and consequently sent to the landfill each week. CJs agreed to allow a separate trash container be placed behind their building and committed their shuckers to dumping the wa ste shells into the bin. After several weeks, the bin was not very full. Upon observing the shuckers at work, it was noticed that in the fast-pace of their shucking, they often forgot to separat e the shells from the regular trash. As a result, accumulating a large enough quantity of shells for testing was a long process. One technique the author tried that was su ccessful was to collect the shells in person at the restaurant as the patrons and t he shuckers were finished. This was time consuming, but having a presence at t he restaurant seemed to help immensely. Cleaning One of the goals of this research was to make as little environmental impact as possible. With that in mind, several methods were tested in order to effectively clean the shells. The shells were first rinsed thoroughly t hen laid out in the sun for several days. This method was not very effective, producin g an extremely unpleasant odor, as well as ineffectively removing remaining oyster meat and miscellaneous food items such as cocktail sauce. The shells were then soaked in a solution of bleach water for several days. Though this was more effective in removing so me of the odor, it did not entirely remove

PAGE 33

33 the meat and miscellanea. The shells were then tumbled in a concrete mixer with bleach, water and different abrasives such as sand, #89 pea gravel and #67 limestone. Of all the combinations attempted, using two shovels of pea gravel with the bleach water did prove reasonably effective in cleani ng the shells; however, it did not adhere to the environmental standar ds of the project. A different batch of oyster shells were first thoroughly rinsed in clean water then placed in a concrete mixer with a strong water and white vinegar solution, as well as two shovels of pea gravel. This mixture was allowed to turn for 30-45 minutes. The pea gravel was then sifted out, the shells rinsed off and then left to soak for approximately 48 hours in a baking soda bath. The resultin g shells had no meat or miscellanea and little or no odor. A B Figure 3-1. Shell cleaning. A) Fresh shells were mixed for approx imately 30-45 minutes in a mixture of pea gravel and vinegar. B) Once rinsed clean of vinegar and gravel, shells were left to sit fo r 48 hours in a baking soda bath. Crushing The shells were required to pass between the and #8 sieves, this would require the shells to be crushed. Severa l methods were used to crush the shells including mechanical and manual.

PAGE 34

34 The clean, dry shells were taken to the local Florida Department of Transportation materials testing facility and run through an adjustable mechanical jaw crusher. The shells were dropped through a top slot and pressed against an iron plate with a separate movable plate. The resulting crushed shells fall into a collection bin. However, because of the flat nature of the shells, it was found that they often slip past or between the movable plate. This result ed in whole or nearly whole shells in the collection bin. It was also found that, unlik e limestone that crushes into smaller rocks, oyster shells tend to flake and shear resulting in shells that were their original length and width but not depth. Many of the shells had to be run through several times to achieve a small enough size. Figure 3-2. Crushing shells at the Flor ida Department of Transportation Materials Research Lab. There were several occasions when it wasnt possible to access the DOT machine. Hand tools such as hammers, sledgehammers and myriad other heavy objects found around the Rinker lab were used to crush the shells. The clean, dry shells were placed between layers of plastic for containment then pounded with the various tools. This did crush the shells, though at a much slower rate. Often, the sledgehammer

PAGE 35

35 would pulverize the shells or not break them into small enough pieces, and often did both at the same time. Obviously, this is a very time consuming process and was not capable of producing a large qua ntity of shells, but did suffice for providing enough shells for testing purposes. The only other process to mass crush she lls that was tried was to run them over with a Caterpillar 226B Skid Steer Loader. Clean, dry shells were placed in two rows, one for each side of the tractor, on a conc rete driveway. The tractor then repeatedly drove over the shells, using it s weight of 5,800lbs to crush them with the tires. After the shells were collected, a large amount of foreign matter was found to be mixed in, apparently from the drivew ay and the tractor tires. These shells had to be well rinsed and sifted to remove all debris. A B Figure 3-3. Using a Caterpillar Skid Steer for oyster shell crushing. A) Oyster shells on bare concrete. B) Oyster shells encased in thick plastic to prevent contamination. The larger shells were sieved out, cleaned and dried then placed between very thick layers of plastic. Once again the trac tor was used to repeatedly run over the shells. The tractors tires were sprayed down with wate r to prevent any dirt or debris that may

PAGE 36

36 penetrate through the plastic. This time bec ause the shells were protected by plastic and the tractor tires had been rinsed of surface dirt, they remained clean. Pervious Concrete Mix Design Introduction Over the course of this research, seve ral methods to attain a pervious oyster shell concrete mix design were implemented. During the first stages, less traditional techniques were used. The later phases follow ed more conventional practices. It should be noted that there are very few ack nowledged standards fo r the development and testing of pervious concrete from national entities such as the American Concrete Institute (ACI) or t he American Society for Testing and Materials (ASTM). As a result, many methods used are either based on simila r concrete standards (such as lightweight concrete), or are the result of scientific hypothesis. Sieve Analysis Sieve analyses were run to determine t he particle size distribution of each specific aggregate. Depending on what was desired for a mix design, particular sieves may be added or removed to accurately assess what range was needed. Testing was performed in accordance with the ASTM C136 procedure for coarse aggregate. To achieve an accurate and diverse sample of oyster shells, the crushed pieces were spread evenly into a large oval onto a concrete surface that had been cleaned of all debris. The shells were then divided into quar ters then opposite sections removed. The remaining sections were mixed and the procedure repeated until approximately 2000g remained. The shells we re then placed into a mechanical sieve shaker, which held a designated nest of siev es. The shaker was run for 10 minutes, allowing sufficient opportunity for the shells to reach the bottom. Over the course of this

PAGE 37

37 research, several sieve analyses were run and different sieves were added or removed from the nest to accommodate large or smaller openings. After the shaking was complete, individual sieves were weighed to determine how many shells remained in each. From these calculations a particle size distribution and fineness modulus was determined. Unit Weight Testing for the unit weight of crushed oyster shell was done in accordance with the ASTM C29 procedure. A random sample of shell was oven dried to 100C then cooled to room temperature. The weight of the measure was taken and the volume of the measure was confirmed. To determine th e loose unit weight of the shell, the measure was filled to overflowing with a scoop and the surface was leveled with a straightedge. The weight of the measure with the shell wa s taken, and then the weight of the measure alone was recorded. To det ermine the compact rodded unit weight, the measure was filled one-third full and leveled with the fingers. The layer was then rodded 25 times with a 5/8 diameter tamping rod ev enly across the surface. This process was repeated two more times until the measure is filled to overflowing. It should be noted that when rodding each layer, the tamping rod should not be allowed to penetrate into the previous layer. After the final rodding processes, the surface was leveled with a straightedge. The weight of the measure with the shell wa s taken, and then the weight of the measure alone was recorded. Specific Gravity Testing for the specific gravity of crushed oyster shell was done in accordance with the ASTM C127 procedur es. A random sample of shell was selected in a quartering process similar to that done in the sieve analysis and run through the

PAGE 38

38 mechanical sieve machine where everythi ng beneath the # 4 sieve was discarded. The sample, totaling 3000g, was s ubmersed for 24 hours. After re moval from the water, the sample was placed on an absorbent towel and blo tted dry, only to the point that that any visible water was removed. This condition is called Saturated Surface Dry (SSD). The SSD shells were weighed and then placed into a test basket. The basket was then fully submerged in water, having taken care to re move all entrapped air that might be in the shells. The weight of the s hells under water was then dete rmined. The shells were then removed from the water and placed into a constant temperat ure oven to dry. The Ball Test Perhaps one of the most simple, yet im portant tests to determine a concrete mixs water content is called the ball test. Not to be confused with the Kelly Ball test, referenced in ASTM C360, the pervious conc rete ball test requires no machines or apparatus to perform. The researcher simply attempts to form a ball with a handful of concrete. If the mix is too dry and difficult to compact and cure, the ball will crumble similar to blue cheese; too wet and the paste can slump off and the voids will close. Very small batches were made based on a mix having a water-to-cement ratio (W/C) of 0.33 and an aggregate-to-cement ratio (A /C) of 4.0. Incremental additions of sand and water were made to the initial mi x and the ball test was performed with each small batch. Several batches were also made with varying gradations of shell sizes. The final mix design, dubbed Mix 8 is found in Table 3-1. Table 3-1. Pervious oyster shell concrete. Mix 8 Ratio Water/Cement (W/C) 0.35 Aggregate/Cement (A/C) 4.0 Admixtures 0

PAGE 39

39 Mixing and Compaction Techniques Introduction Once a mix design was decided upon, se veral batches were made with different compaction techniques. Testing was perform ed on samples derived from batches made in three separate trials; two batches were placed in conventional waxed cardboard molds and one was placed in a 4x4.5 mold. All batches were removed from their respective molds at appropriate times and allowed to cure in the controlled environment of the Rinker laboratory for one year. Due to uncontrollable events, not all samples were able to be run through the full course of test ing. Of the one-year samples, two Standard Proctor compaction, four Modified Procto r compaction and one non-compacted samples were put under compressive testing. Three Standard Proctor compaction, four Modified Proctor and two non-compacted samples were tested for permeability. Only one Standard Proctor, three Modified Proctor and one non-compacted samples were tested in the water displacement procedure. O ne-each of the Standard Proctor, Modified Proctor and non-compacted samples placed in the 4x4.5 molds were tested for permeability, however only the Standard and Modified compacted samples were capped and tested for compression and none were tested for density. One 6x6 sample, placed with the r odded compaction technique, was made 21 days before testing. This sample was tested for compression and permeability. A large batch of pervious oyster she ll concrete was mixed seven days before testing. From this mix, three samples each using the rodded, vibrated and Standard Proctor compaction techniques were placed in 4x8 waxed cardboard cylinders. One sample each using the vibrated and Standard Proctor compaction technique was placed in a 6x6 waxed cardboard cylinder.

PAGE 40

40 Mixing Several mixing techniques were employed during the course of this research depending on the batch size needed for testing. For the sma llest batches, a clean, dry 6x12 plastic testing cylinder was used as the mixing container. In accordance with ASTM C192 7.1.3 for hand mixing, the she lls and cement were thoroughly combined then the water was slowly added until the mass achieved a desired texture and appearance and the shells were uniformly coated. The concrete was then mixed for three minutes, allowed to re st for three minutes, then mixed for an additional two minutes. For slightly bigger batches, a clean and dry five-gallon bucket was used as a mixing container and the same procedure listed above was followed. For the largest batch run, a proc edure devised by a Southern Illinois Civil Engineering research group st udying no-fines concrete was successfully used. This batch was mixed in a 1/3 yd machine mixer. Prior to running the machine, the mixer was buttered with a small am ount of water; shells were then added to the mixer along with one-third of the mixing water. The mixer was then run for five minutes. The Portland cement was added and the mixer was r un for another five minutes. Finally, the remaining water was added and the mixer was run for a final five minutes. The total mixing time was 15 minutes (Ghafoori, 1995). Compaction Several methods were employed to derive the best means of consolidating the pervious oyster shell concrete. Compaction is necessary to form a strong bond between the paste and aggregate and also to provide a smooth surface. Because of the low water content, pervious concretes mixes must be placed and compacted very quickly.

PAGE 41

41 And, unlike conventional concretes mixes, commonly used equipment such as a bullfloat may not be employed as there is ve ry little excess water to lose. Improperly compacting the concrete could potentially cause void collapse, therefore reducing its permeability or increase its likelihood of spalling once the concrete has cured. The most common commercial compaction techniques used to date are the vibratory screed or the weig hted roller, however these methods are not able to be verified or regulated by any ASTM standard or recreated on such a small scale so they were not tested in this research. At the onset of this research, ther e were no beam molds available so other means were sought to test in horizontal app lication. As a resul t, early testing was performed in simple kitchen storage contai ners. Later research was performed in standard 4x8 or 6x6 vertical waxed cardboard cylinders. These vertical molds were arguably not the best choice for th is type of aggregate or concrete. Testing performed with the kitchen stor age containers used rather rudimentary means of compaction including surface skree ing and simply shaking the container as an alternate to vibration. When using waxed cardboard cylinders, f our methods were used to compact the pervious oyster shell concrete: rodding, vibration, Standard Proctor and Modified Proctor. Test cylinders were also made th at had no compaction applied. These are referred to as non-compacted. For these samples, the concrete was placed into the containers, screed to a level finish then co vered. Non-compacted samples imitate a

PAGE 42

42 real-world scenario where the concrete is just placed and raked into position, screed smooth then left to dry. Figure 3-4. Early molding techniques. Rodding In accordance with ASTM C192 7.4.2 Methods of Consolidation, Rodding, testing cylinders 2-5 in diameter require a rod of 3/8 in diameter, and cylinders 6 require a 5/8 rod therefore the appropriate rods were used in equ al strokes of 25 per layer, for two layers. The strokes were spread uniform ly over the specimen and, on the upper layer penetrated one inch into the bottom layer. After each rodded session, the sides of the cylinder were tamped 10-15 times. It was found that the fissile nature of the oyster she ll caused them to seemingly lock together with the insertion of the rod, making it very diffi cult to penetrate deeply into the mold and at times impossible to fully reach the lower level. Vibration In an attempt to create as close to a real-world scenario as possible, yet still maintain a standardized testing methodology, specimens were compacted on a 75-Hz

PAGE 43

43 frequency external vibrator but with additional weight added on top. Larger 6x6 molds were weighed with 8.6kgs and 4x8 molds were weighed with 3kgs. Specimens were slightly overfilled, weighed down, and then submitted to uniform vibration until the concrete ceased to settle. At that point the concrete was screed to an even finish, covered with plastic and carefully removed from the vibrator. Figure 3-5. Vibration of a 6x6 sample. The Proctor Method Most commonly used by scientists for so ils research but for our purposes an effective tool for achieving high density is the Proctor test. The original Proctor test, ASTM D698, uses a 4-inch diameter mold that holds 1/30th ft of material, and calls for compaction in three layers of 25 blows by a 5.5 lb hammer, which falls 12 inches. This has a compactive effort of 12,400 ft-lb/ft. The "Modified Proctor" test, ASTM D1557, uses the same mold, but uses a 10 lb. hammer falling for 18 inches, with 25 blows for a compactive effort of about 56,000 ft-lb/ft. This procedure was followed exactly for three batches of concrete, however it was discovered that the concrete developed ve ry distinct striations at the layers. Several samples broke along the striations with the application of hand twisting or

PAGE 44

44 pulling. Therefore it was deci ded that samples made in the fourth and final batch would be placed in only one layer. With that except ion, ASTM D698 procedures were followed. In this final batch, the collar from a Proctor hammer measure was placed on top of a waxed cardboard cylinder. A weight or flat surface was placed upon the surface of the concrete and the Proctor hammer was appl ied to that surface. This method was chosen to help distribute the force of the hammers impact, to reduce the likelihood of localized surface compaction that seem to be common with the application of the Proctor hammer, as well as to protect the concrete from developing harsh striations, which were noted above. The collar prevented the weight and concrete from displacement. Figure 3-6. The customized St andard Proctor compaction method. Pervious Concrete Testing Unit Weight for Fresh Concrete There is a large amount of debate in t he world of concrete about the correct testing method for determining the unit weig ht of pervious concrete. Currently, the American Concrete Institute (ACI) requi res testing using the rodding and tamping techniques for measuring density found in ASTM C138, as well as the jigging technique

PAGE 45

45 in ASTM C29. However, this requirement will soon change because of a recent amendment to standards. In October of 2008, ASTM released its C1688 Standard Test Method for Density and Void Content of Fres hly Mixed Pervious Concrete. The new test entails a sample measure to be filled in tw o layers, with each layer compacted by 20 blows of a Standard Proctor hammer. Accordi ng to the managing dire ctor of research and materials engineering at the National Ready Mixed Concrete Association and chairman of ASTM Subcommittee C09.49, th is new method should more accurately represent consolidation results fo und in the field (Palmer, 2009). At the onset of this research, ASTM C1688 had not been released and ASTM C138 was the only available unit weight test. Therefore, for the purpose of consistency, all unit weight tests in this research were performed in accordance with ASTM C138. To perform the unit weight test, an empty measure is weighed then filled in three layers of approximately equal volume. Ea ch layer is rodded 25 times with a 5/8 diameter tamping rod. After rodding, the side of the m easure is tapped with a rubber mallet 15 times, which releases any trapped air bubbles in the concrete. After this process is concluded, the t op of the concrete is screed o ff with a sawing motion and the measure is weighed again. Water Displacement Test It is generally accepted that for any given mixture proportion of pervious concrete, strength and permeability are a function of the conc retes density. The greater compaction effort, the higher the strength and lower the perm eability rate (Obla, 2007). For these reasons it is vital to test the density of the various mix designs and compaction efforts for any concrete. For the purposes of this research, a test based on the Archimedes theory of water displacement was developed.

PAGE 46

46 Water was placed into a contained up to a given mark. The concrete sample was submerged in the water and the amount of water displaced was determined. Permeability Constant Head Testing for permeability in 4x8 sa mples was done using a steady stream of water from a simple water faucet. Samples were either retained or returned to waxed cardboard cylinders whose top and bottoms had been removed. Above and beneath these cylinders was placed the collar of a Pr octor Method container. High-adhesive tape was wrapped several times around the two to ensure a good seal. Water was allowed to flow freely in a steady, fullhead stream from appr oximately 5 above the sample for several minutes for conditioning. When a 1 head of water could be maintained in the collar, a collection bucket was placed undern eath the sample and the amount of water collected in a given amount of time was deduced. It must be noted that severa l of the samples did not maintain a 1 head of water; at times could be achieved, but often none at all. A B Figure 3-7. Permeability test ing. A) Waxed cardboard Cylin der sample during overhead flow permeability test. B) Sample plac ed in 4x4.5 mold during overhead flow permeability test.

PAGE 47

47 Permeability Falling Head Apparatus Perhaps the most accurate way to test permeability of pervious concrete is with a falling head apparatus. Unfortunately, only three pervious oyster shell concrete samples were large enough to fit into the available machine. To use the apparatus, 6x6 samples are fi tted into a latex membrane with a clear graduated plastic cylinder placed on top. Water is poured into the plastic cylinder and allowed to drain through the concrete specim en until it is at the same level in the graduated cylinder as it is in the drain pipe. Between the plastic cylinder and the drain pipe is a control valve. This valve is opened a nd the time for water to fall from a given point in the graduated cylinder to a final head is taken. The rate of permeability is found using the calculations for Darcys Law: (3-1) Where: = Inside diameter of falling head tube (Length) = Inside diameter of parameter (L) L = Sample length = Initial hydraulic head in falling head tube (L) h = Final hydraulic head in falling head tube (L) t = Time it takes to change from to h (Time)

PAGE 48

48 Figure 3-8. Concrete specimen inside the falling head permeability apparatus. Sulfur Capping Several hours prior to testing, each of the specimens was transported to the FDOT testing facility and capped with a high strength sulfur-based capping compound. This capping procedure was performed to assure even loading across the top and bottom faces of the cylinder. The capping was done in accordance with procedures outlined in ASTM C 617. Figure 3-9. Sulfur caps being applied to the pervious oyster shell concrete samples.

PAGE 49

49 Compression Testing After capping, samples were subjected to compression testing. Eleven samples were tested at seven days old, one samp le was tested at 21 days and seven were tested at 14 months. It must be noted that several samples broke prior to compression testing. These breaks seemed to be a direct result of the concre te being placed in layers. So as to not sacrifice any pot ential knowledge that could be gained from compression testing the samples, the cylinders were capped at their remaining height. A. B. Figure 3-10. Compression testing. A) M odified Proctor sample. B) Vibrated sample.

PAGE 50

50 CHAPTER 4 DATA ANALYSIS AND RESULTS Aggregate Analysis The fact that oyster shell would prove to act as a very different form of aggregate than limestone (the most common aggregate in pervious concrete) was first made clear when the shells were crushed. Unlike lime stone, which generally breaks into fairly consistently-sized rounded, smaller rocks, oyster shells often flake, resulting in pieces retaining their original size just smaller in depth. It should also be noted that this flaking, fissile tendency caused the crushed shell to present in longer, flatter pieces of wi der gradations, but all with similar flake-like structures. When amassed, thes e pieces of crushed shell were very difficult to separate and scoop for testing. The scoop would almost glide over the shells unless an opening was forced. This same interlocking-type action was later noticed in the rodded compaction technique. Once the shells were crushed, thei r average gradation was found through sieve analyses. Several tests were run with and without the inclusion of sand. It was found that the crushed shells alr eady contained a large amount of fines, which would make the concrete less permeable. Finer sieves we re removed from the nest to eliminate a larger amount of smaller aggregate. It was also discovered that the average random crushed sample contained a high number of larger shells, which would cause the concrete to break apart or necessitate a hi gher aggregate/cement rati o. It was finally decided to keep all shells passing through t he sieve and reject those falling beneath the #8 sieve (-1/2 + #8). The results of the sieve analysis may be found in Appendix A.

PAGE 51

51 ASTM tests for rodded compact unit weig ht, specific gravity, density and absorption were performed, as well as a water displacement test to confirm density and specific gravity. The resu lts are found in Table 4-1. Table 4-1. Properties of the Eastern Oyster Shell. Eastern Oyster Shell Specific Gravity dry bulk Specific Gravity SSD Specific Gravityapparent Density dry bulk (lb/ft) Density SSD (lb/ft) Density apparent (lb/ft) Absorption (%) Unit Weight compact (lb/ft) Void Content (%)3.04 3.17 3.49 189.79 197.28 217.26 4.2 65.7 65 Identify a Viable Concrete Mix Design Early Trials The first several mix designs relied heavily on the inclusion of sand, as well as utilizing large-size shells. These mixes were also placed into plastic kitchen storage containers that were often stored outside for 24-hours cove red then were uncovered or were stored outside and never covered at a ll. Dry samples generally presented a solid impervious bottom or were simply a solid concrete substance. Unfortunately, research data was not very well documented except for photographic evidence. None of the early sa mples were ever subjected to standard ASTM testing procedures. However, knowle dge was gained from t hese mixtures that showed the importance of a low water/cement ratio, a low aggregate/cement ratio, as well as the discretionary use of fines in pervious concrete. The Ball Test Incrementally adding sand and water to a basic blueprint mixture was key in finalizing the ultimate mix design. Very sm all batches were made based on a mix having

PAGE 52

52 a water/cement ratio (W/C) of 0.33 and an aggregate cement ratio (A/C) of 4.0. The goal was to have a ball of concrete form in the researchers hand. If the mix was too soupy it would fall apart and lose permeability ; too dry and it woul d crumble. It was found that mixtures with a co mplete range of shells had very good adhesion with a W/C of 0.35 and an A/C of 4. 0; however the larger shells s eemed to make the ball fall apart. The same mix was made with a more defined range of shells and the ball retained its shape. The basic mix was made with the inclus ion of 100g of sand. This made a very well-formed ball. However, upon drying it was found that this ball had very little permeability. This same mix was also m ade with 100g of sand and the addition of 75g of water. This mixture had a large slump and would not form a ball. It was finally decided that the initial mi x, without any addition of fines and with a narrowed selection of shell size was the best c hoice for a mix design. Overall, this test was a very good indicator of what effect small changes can have on mix designs. Pervious Concrete Testing Void Content Common air-void factors for pervious c oncrete are listed as 15-30% (Huffman, 2005). When the pervious oyster shell concrete was tested for density, however, its airvoid ratio proved to be much higher, with a range of 26-58%. The average void ratio for the week-old rodded compaction concrete was 48.2%, the Standard Proctor compaction concrete was even higher at 53.9% and t he vibrated compaction concrete had a void ratio average of 51.8%. The year-old Modi fied Proctor compacted concrete showed much more acceptable results with an av erage air-void ratio of 28.5%. While an average cannot be taken of the year-o ld Standard Proctor compaction and non-

PAGE 53

53 compacted concrete samples because only o ne sample was able to be tested, their results were 37.7% and 58.6%, respectively A graph charting the relationship between porosity and the different compaction te chniques is found in Figure 4-1. Figure 4-1. Relationship bet ween porosity and compaction. Table 4-2. Specific Gravity and porosity by compaction method on samples placed in 4x8 molds. Compaction Method 7-day 1-year Specific Gravity Porosity (%) Specific Gravity Porosity (%) Rodded 1 0.87 47 2 0.86 47 3 1.03 51 Vibrated 1 1.08 50 2 1.08 53 3 1.09 52 Std. Proctor 1 1.13 55 0.74 38 2 0.97 49 3 1.15 57 Mod. Proctor 1 0.5 31 2 0.42 26 3 0.44 29 Compression Pervious concrete is not known for its strength in compression. According to the Web site www.perviouspavement.org a holding of the American Concrete Insti tute, the

PAGE 54

54 average range of compressive strength for pervious concrete is between 500 to 4000 psi. Pervious oyster shell concretes strength tested on the very low end of that spectrum. When testing the 4x8 cylinders, the aver age seven-day compressive strength for the rodded compaction concrete was 445 psi. The Standard Proctor method compacted concrete broke at 453 psi and the vibration compacted concrete had a very low compressive strength of 269 psi. The one-year Modified Proctor method cylinders tested at a much stronger average of 1071 ps i, though the average one-year Standard Proctor method samples broke at only 251 psi. It must be noted that there were only two Standard method cylinders tested and one was broken before testing and consequently sulfur capped at partial height. This did not seem to have much of an effect on the breaking strength. The one-year non-co mpacted cylinder failed at 387 psi. The highest compressive strength result s resulted from break ing the one-year Standard and Modified Proctor method concre te samples that were placed in the 4x4.5 molds. The Standard Proctor cylinder failed at 1552 psi and the Modified Proctor broke at 1592 psi. The 21-day-old vibration compacted concrete sample broke at 481 psi. Both the rodded and Standard Proctor method compacted concrete samples were at seven-day strengths. The rodded sample broke at 505 psi while the Standard Proctor sample had a compressive strength of 543 psi. A graph charting the relationship between compressive strength and the different compaction techniques is found in figure 4-2.

PAGE 55

55 Figure 4-2. Relationship between co mpressive strength and compaction. Permeability One of the best indicators of a pervious concretes efficacy is its permeability. In its document Pervious Concrete ACI 522R-06, ACI indicates that the average drainage rate of a pervious concrete will gener ally fall between 2 to 18 gal/min/ft, and it also state that this rate will vary depending on the aggregate us e and density of the mixture. All one-week compaction efforts resulted in drainage rates well above that mentioned by ACI. The one-year samples test ed with a wide range of results, from highly permeable to almost impervious. The average permeability of a vibration compacted 6x6 cylinder tested with the falling head parameter apparatus was 27gal/min/ft, while the average permeability of a vibration compacted 4x8 cy linder tested by overhead flow was 22.6 gal/min/ft. A closer similarity in averages was found with the rodded compaction samples, testing 25.7 gal/min/ft in the 6x6 cylinders and 24 gal/min/ft in the 4x8 cylinders. Standard Proctor method compaction samples showed an even average with both testing methods draining at 21 gal/min/ft in both 6x6 and 4x8 cylinders.

PAGE 56

56 The widest range of permeability was found in the one-year samples. The samples placed with Standard Proctor meth od compaction had an average permeability of 9 gal/min/ft, while those placed with t he Modified Proctor method had an average of 0.5 gal/min/ft. The non-compacted samples dr ained conversely at an average of 21.7 gal/min/ft. Those samples placed in the 4x4.5 molds were tested in situ. The Standard Proctor compaction method sample had a drai nage rate of 7.3 gal/min/ft, while the Modified Proctor compaction method sample drained at 2 gal/min/ft and the noncompacted sample at 18.9 gal/min/ft. Figure 4-3. Relationship betw een permeability and compaction. Table 4-3. Compressive strengths and per meability rates of one-year samples in 4x4.5 molds. Compaction Method Compressive Strengt h (psi) Permeability (gal/min/ft) Std. Proctor 1552 7 Mod. Proctor 1596 2 Non-Compacted N/A 19

PAGE 57

57 Table 4-4. Compressive strengths and per meability rates of 7-day and 21-day samples placed in 6x6 molds. 7-day 21-day Compaction Method Compressive Strength (psi) Permeability (gal/min/ft) Compressive Strength (psi) Permeability (gal/min/ft) Rodded 505 26 Custom Std. Proctor 543 21 Vibrated 481 27 Table 4-5. Compressive strengths and perm eability rates of 7-day and 1-year samples placed in 4x8 molds. 7-day 1-year Compaction Method Compressive Strength (psi) Permeability (gal/min/ft) Compressive Strength (psi) Permeability (gal/min/ft) Rodded 1 460 24 2 516 25 3 360 23 Std. Proctor 1 442 21 297 10 2 466 20 205 9 3 451 22 Vibrated1 N/A 23 2 206 22 3 329 23 Modified Proctor 1 920 0.3 2 1291 0.3 3 1093 0.9 4 978 N/A Non-Compacted 378 21

PAGE 58

58 CHAPTER 5 SUMMARY AND CONCLUSION The purpose of this re search was to develop a pervious concrete using recycled oyster shells as the aggregate. To accomp lish this, the following tasks were performed: Eastern Oysters shells were collected, cleaned and crushed, then their properties analyzed for particle size, specific gravity, density, absorption and unit weight. A pervious oyster shell concrete mix was designed using trial batches and the ball test. The pervious oyster shell concrete was placed with three compaction techniques then tested for unit weight, density, permeability and compression. It was found that a pervious oyster she ll concrete could be designed, however it was also discovered that the process of ac quisition and cleaning of the shells may prove too labor intensive to make the mix viable on a large scale. There were also complications in the test met hods implemented in this research that might have led to a weakness in the overall data. Complications There was some inconsistency in data acquis ition. Some samples were not included in a few of the tests so broad scope conclusions couldnt be drawn. A larger quantity of samples should have been made so a better scientific data set could be taken. Based on the research, several assertions may be concluded regarding the pervious oyster shell concrete, as well as the oyster shell as an aggregate. Conclusions It may be assumed that wit h the proper compaction tec hnique, the oyster shell will be proven to be a very st rong aggregate. This presum ption is based on the locking tendency of the aggregate shell that was observed in the r odding compaction technique

PAGE 59

59 as well as when the shell was amassed. The validity of this assumption needs to be established with further research. The conclusion may be drawn that pervious oyster shell concrete should be limited to outside use, as a faint odor persi sts around the material even after curing for one year. This issue may be resolved if the shells were superheated or cleaned with other methods, however that is a topic t hat would need to be va lidated with additional research. Recommendations for future research: Place pervious oyster shell concrete into large panels and compact with several different techniques, namely a wei ghted roller and vibrating roller. Test pervious oyster shell concrete for permeability while in the above-mentioned panels. Several methods for in-situ test ing are currently being developed for standardization and are available for acquisition via the Internet. Develop additional mix designs that in corporate recycled content such as blast furnace slag or fly ash, to further increas e not only the physical properties of the concrete, but also the sustai nable attributes as well. Develop additional mix designs that in corporate different gradations of shell aggregate in order to potentially provide better compaction ratings.

PAGE 60

60 APPENDIX RESULTS OF OYSTER SHELL SIEVE ANALYSIS Figure A-1. Results of test using sieve sizes 1-#30.

PAGE 61

61 LIST OF REFERENCES Adams, D. (2007). Tabby: The Cement of the Lowcount ry < http://www.beaufortcountylibrary .org/htdocs-sirsi/tabby.htm > Sept. 2005 American Concrete Institute (2006). Pervious Concrete ACI R-06. American Concrete Institute, Farmington Hills, MI. Chiras, D. (2002). Natural Resou rce Conservation Prentice Hall, Inc. Upper Saddle River, New Jersey Deagan. K and Cruxent, J. (2002). Columbuss Outpost Among the Tainos Edwards Brothers, Inc. USA Ghafoori, N. and Dutta S. ( 1995). Laboratory Investigati on of Compacted No-Fines Concrete for Paving Materials, Journal of Materials in Civil Engineering Hitzman, M., Dilles, J., Barton, M., and Boland, M. (2009). < http://www.geosociety.org/gsatoday/archiv e/19/8/pdf/i1052-5173-19-8-26.pdf > Sept. 2009 Huffman, D. (2005). Understanding Pervious Concrete The Construction Specifier. Kenilworth Media, Richmond Hill, ON. Kibert, C. (2005). Sustainable Construction John Wiley & Sons, Inc., Hoboken, New Jersey. Kosmatka, S. Kerkhoff, B. and Panarese, W. (2002). Design and Control of Concrete Mixtures Portland Cement Association, Skokie, IL. Kresge, P. (2009). Pervious Concrete FAQs < http://findarticles.com/p/articles/mi _hb5123/is_200710/ai_n3224 6224/?tag= content;col 1 > Sept. 2009 Low, M. (2005). Material Flow Analysis of Concrete in the United States. MS Thesis, Massachusetts Institute of Technology, Cambridge, Mass. Matos, G. (2009). Use of Minerals and Ma terials in the United States From 1900 Through 2006, Fact Sheet 2009-3008. U.S. Geological Survey, Reston, VA. Michigan Technological University (2003). Mining Glass from the Waste Stream < http://www.imp.mtu.edu/information/wgpc.html > Oct. 2009 Obla,K. (2007). Pervi ous Concrete for Sustainable Development< http://www.nrmca.org/research/P ervious %20recent%20advances%20in %20concrete%20technology0707.pdf > Aug. 2009

PAGE 62

62 Offenberg, M. (2008). Is Pervious Concrete Ready for Structural Applications? < http://www.structuremag.org/ article.asp x?articleID=532 > Sept 2009 Palmer, W. (2009). Testing Pervious Concrete < http://www.concreteconstruction.net/industrynews.asp?sectionID=985&articleID=936324 > Oc t. 2009 Renn, D.F. (2009). The Keep of Wareham Castle < http://ads.ahds.ac.uk/catalogue/adsdata/arch-7691/ahds/diss emination/pdf/vol04/4_056_068.pdf > Sept. 2009 Sheehan, M., Sickels-Taves,L. and Bjork, J. (1998). The Lime Middens of Cumberland Island < http://crm.cr.nps.gov/archive/21-9/21-9-o3.pdf > Jan. 2007 Shulman, D. (2005). S ynthetic Aggregate < http://www.pitandquarry.com/pitandquarry/content/printContentPopup.jsp?id =172922 > Sept. 2009 Sickels-Taves, L. (1996). Southe rn Coastal Lime Burning < http://crm.cr.nps.gov/a rchive/19-1/19-1-8.pdf >Jan. 2007 Sickels-Taves, L. (1998). Handle with Care Tabby is No Ordinary Concrete, < http://www.gashpo.org/assets/documents/tabby_scan ned.pdf > Sept. 2007 Sickles-Taves, L. (2008). The Care & Preservation of Historic Tabby < http://thehenryford.org/re search/caring/tabby.aspx > Aug. 2009 Sickles-Taves, L. and Sheenan, M. (1999). The Lost Art of Tabby Redefined Architectural Conserv ation Press, Southfield, MI. Stone, R. (1894). State Laws Relating to the Management of Roads Bulletin No.1. U.S. Department of Agricu lture, Washington, D.C. Sullivan, B., (1998). Tabby: A Historical Perspective of an Antebellum Building Maerial in McIntosh County, GA, < http://www.gashpo.org/assets/documents/tabby_scan ned.pdf > July 2005 United States Environmental Protection Agency (2008). Municipal Solid Waste Generation, Recycling, and Disposal in the Un ited States: Facts and Figures for 2007, Washington, D.C. Wilburn, D. and Goonan, T. (1998). Structure of the Aggregates Industry < http://pubs.usgs.gov/c irc/1998/c1176/c1176.pdf#6 > Aug. 2009

PAGE 63

63 Wilson, J. (2008).Guiding Pr inciples for Sustainability < http://www.nssga.org/sustainab ility/pdfs/NSG1578Su stainWeb.pdf > National Stone Sand & Gravel Association. Sept. 2009 WOX Info (2009). Tracing the Maritime Silk Road, < http://www.whatsonxiamen.co m/ifm_infobank.php?t itleid=507 >ct. 2009

PAGE 64

64 BIOGRAPHICAL SKETCH Kristy Noel Kelley is an eighth generat ion North Floridian born in Tallahassee in 1974. She is an alumna of Tallahassee Community College where she took an Associate in Arts and graduated Cum Laude, and Florida State University, where she took a degree in English with a minor in history. After working several years as a writer with the Tallahassee Democrat, she worked for the Tallahassee Trust for Historic Preservation (TTHP). It was working for t he TTHP that she began to see her plan for working with historic buildings unfold. Kell ey worked as a laborer for a short time for Lambert Construction Company in Tallahass ee. It was there that her life changed, discovering the fragility of bones and ligamen ts. She entered the M.E. Rinker, Sr. School of Building Construction in August of 2004 and has pursued her graduate degree despite many setbacks and frustrations, with the support of her incredible family. She will graduate on her 35th birt hday in December of 2009.