Full report

Original page 1

NASA/TM—2008–215250 Sulfur ‘Concrete’ for Lunar Applications– Environmental Considerations R.N. Grugel Marshall Space Flight Center, Marshall Space Flight Center, Alabama February 2008

Original page 2

The NASA STI Program…in Profile Since its founding, NASA has been dedicated to the advancement of aeronautics and space science. The NASA Scientific and Technical Information (STI) Program Office plays a key part in helping NASA maintain this important role. The NASA STI program operates under the auspices of the Agency Chief Information Officer. It collects, organizes, provides for archiving, and disseminates NASA’s STI. The NASA STI program provides access to the NASA Aeronautics and Space Database and its public interface, the NASA Technical Report Server, thus providing one of the largest collections of aeronautical and space science STI in the world. Results are published in both non- NASA channels and by NASA in the NASA STI Report Series, which includes the following report types: • TECHNICAL PUBLICATION. Reports of completed research or a major significant phase of research that present the results of NASA programs and include extensive data or theoretical analysis. Includes compilations of significant scientific and technical data and information deemed to be of continuing reference value. NASA’s counterpart of peerreviewed formal professional papers but has less stringent limitations on manuscript length and extent of graphic presentations. • TECHNICAL MEMORANDUM. Scientific and technical findings that are preliminary or of specialized interest, e.g., quick release reports, working papers, and bibliographies that contain minimal annotation. Does not contain extensive analysis. • CONTRACTOR REPORT. Scientific and technical findings by NASA-sponsored contractors and grantees. • CONFERENCE PUBLICATION. Collected papers from scientific and technical conferences, symposia, seminars, or other meetings sponsored or cosponsored by NASA. • SPECIAL PUBLICATION. Scientific, technical, or historical information from NASA programs, projects, and missions, often concerned with subjects having substantial public interest. • TECHNICAL TRANSLATION. Englishlanguage translations of foreign scientific and technical material pertinent to NASA’s mission. Specialized services also include creating custom thesauri, building customized databases, and organizing and publishing research results. For more information about the NASA STI program, see the following: • Access the NASA STI program home page at http://www.sti.nasa.gov • E-mail your question via the Internet to help@sti.nasa.gov • Fax your question to the NASA STI Help Desk at 301– 621–0134 • Phone the NASA STI Help Desk at 301– 621–0390 • Write to: NASA STI Help Desk NASA Center for AeroSpace Information 7115 Standard Drive Hanover, MD 21076–1320

Original page 3

NASA/TM—2008–215250 Sulfur ‘Concrete’ for Lunar Applications– Environmental Considerations R.N. Grugel Marshall Space Flight Center, Marshall Space Flight Center, Alabama Natonal Aeronautcs and Space Admnstraton Marshall Space Flght Center • MSFC, Alabama 35812 February 2008 

Original page 4

Acknowledgments The author s grateful to Professor H. Toutanj at The Unversty of Alabama n Huntsvlle for hs help and comments and to hs group for provdng sulfur concrete samples. Apprecaton s also expressed to the Marshall Space Flght Center In Stu Fabrcaton and Repar Element of the In Stu Resource Utlzaton (ISRU) as well as Materals & Processes Laboratory EM30 for ther support of ths work. Apprecaton s further expressed to Mr. Curts Bahr for hs techncal expertse n supportng ths work. A sncere apprecaton s also extended to Ms. Lnda Woolf for her crtcal readng of the manuscrpt. The ISRU s a core component of the Vson for Space Exploraton as mplemented by the Scence & Msson Systems (S&MS) Office. The ISRU works to establish, evaluate, and assess the in situ resources available on the Moon and Mars and the technologes needed to utlze and explot these resources. TRADEMARKS Trade names and trademarks are used in this report for identification only. This usage does not constitute an official endorsement, ether expressed or mpled, by the Natonal Aeronautcs and Space Admnstraton. Avalable from: NASA Center for AeroSpace Informaton 7115 Standard Drve Hanover, MD 21076 –1320 301– 621– 0390 Ths report s also avalable n electronc form at <https://www2.st.nasa.gov> 

Original page 5

TABLE OF CONTENTS 1. INTRODUCTION ......................................................................................................................... 1 2. SUBLIMATION CONSIDERATIONS ......................................................................................... 4 2.1 Expermental Procedure .......................................................................................................... 5 2.2 Expermental Results .............................................................................................................. 8 2.3 Dscusson of Sublmaton Concerns ...................................................................................... 14 2.4 Summary of Sublmaton Concerns ........................................................................................ 21 3. EXTREME COLD AND TEMPERATURE CYCLE CONSIDERATIONS REGARDING THE INTEGRITY AND COMPRESSION STRENGTH OF SULFUR CONCRETE ................. 22 3.1 Expermental Procedure .......................................................................................................... 22 3.2 Expermental Results .............................................................................................................. 23 3.3 Dscusson of Extreme Temperature Concerns ....................................................................... 23 3.4 Summary of Extreme Temperature Concerns ......................................................................... 31 4. OVERALL SUMMARY AND CONCLUSIONS ......................................................................... 33 REFERENCES ................................................................................................................................... 34 

Original page 6

LIST OF FIGURES 1. Eght-cubc-yard heater-mxer moble producton unt ...................................................... 1 2. Pourng of Chempruf Sulfur Concrete n an ndustral plant ............................................. 2 3. Chempruf Sulfur Concrete: acd contanment system ........................................................ 2 4. Precast Chempruf Sulfur Concrete tanks for Aqua Rega .................................................. 2 5. Typcal unary sulfur phase dagram ................................................................................... 4 6. A pece of sulfur-based concrete made at The Unversty of Alabama n Huntsvlle ........ 5 7. Photograph of the vacuum chamber used for the sublmaton experments ....................... 6 8. Extrapolaton of the sold-vapor transton lne on a unary sulfur phase dagram ............. 7 9. Comparson of the sulfur concrete surfaces after vacuum processng of the upper rght-hand corner secton ................................................................................ 7 10. Surface morphology of the as-cast pure sulfur ................................................................... 8 11. Surface morphology of the as-cast pure sulfur after vacuum processng on the order of 3–7 × 10–6torr (4–9.3 × 10 Pa) for ≈11 days ......................................... 8 –4 12. Surface morphology of the as-cast pure sulfur after vacuum processng on the order of 3–7 × 10–6torr (4–9.3 × 10 Pa) for ≈54 days ......................................... 9 –4 13. Sulfur - 65 wt. % JSC-1 samples subjected to vacuum processng .................................... 9 14. Sulfur and slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 subjected to vacuum processng ....................................................... 10 15. Mcrograph showng the surface of the as-cast sulfur - 65 wt. % JSC-1 sample ............... 11 16. Mcrograph showng the surface of the sulfur - 65 wt. % JSC-1 sample after 8 days n vacuum ........................................................................................................ 11 17. Mcrograph showng the surface of the sulfur - 65 wt. % JSC-1 sample after 58 days n vacuum ...................................................................................................... 12 v

Original page 7

LIST OF FIGURES (Continued) 18. Mcrograph showng the surface of the as-cast sulfur and slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 sample ................................ 12 19. Mcrograph showng the surface of the slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 sample after 8 days n vacuum .............. 13 20. Mcrograph showng the surface of the slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 sample after 58 days n vacuum ............ 13 ght loss for the exposed surface area (mg/mm2) 21. Plot of the sulfur we as a functon of tme ........................................................................................................... 14 22. Macrograph of the sulfur - 65 wt. % JSC-1 sample after 8 days n vacuum that represents the green diamond datum point, figure 21 .................................................. 22 23. Comparson of measured weght loss for pure sulfur wth the Hertz-Knudsen equaton, αv = 1 .................................................................................................................. 16 24. Comparson of the measured weght loss for pure sulfur and the sulfur concrete samples wth the Hertz-Knudsen equaton, αv = 1, and wth volume fracton, vf, correcton factors ................................................................................................................ 18 25. Photograph of two samples exposed to vacuum for 60 days: left: sulfur - 35 wt. % JSC-1 and rght: 25% sulfur and 20% slca by weght wth 55 wt. % JSC-1 ................................................................................................................... 19 26. Calculated plot showng the effect of temperature on the tme needed to sublmate a 1-cm (0.4-n) thckness of pure sulfur ............................................................................. 19 27. Calculated plot showng the role the ambent pressure plays n regard to the equlbrum vapor pressure ....................................................................................... 20 28. Typcal 2.54-cm (1-n) cube of sulfur concrete used n ths study ..................................... 22 29. A typcal tme temperature plot showng one of the cycles the samples experenced ....... 23 ng compresson testng at –101 °C (–150 °F) ............... 24 30. Photograph of a sample undergo 31. Photograph of crackng exhbted on the surface of sample 22, whch was subjected to 80 cycles between RT (≈21 °C, or 70 °F) and –191 °C (–312 °F).................................. 24 32. Plot of the maxmum compresson strength exhbted by the cycled and noncycled samples ....................................................................................................... 25 v

Original page 8

LIST OF FIGURES (Continued) 33. SEM mcrograph of a fracture surface from noncycled sample 10, a slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1, tested at 21 °C (70 °F) ........................................................................................................ 26 34. SEM mcrograph of a fracture surface from noncycled sample 14, wth the same composition as in figure 33, but tested at –101 °C (–150 °F) ............................................ 26 35. SEM mcrograph of a fracture surface from cycled sample 22, a slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1, tested at 21 °C (70 °F) .................................................................................................................. 27 36. SEM mcrograph of a fracture surface from cycled sample 24, wth the same composition as in figure 35, but tested at –101 °C (–150 °F) ............................................ 27 37. SEM mcrograph of a fracture surface from noncycled sample 4, a sulfur wth 65 wt. % JSC-1 sample tested at 21 °C (70 °F) .................................................................. 28 38. SEM mcrograph of a fracture surface from noncycled sample 6 wth the same composition as in figure 37, but tested at –101 °C (–150 °F) ............................................ 29 39. SEM mcrograph of a fracture surface from cycled sample 18, a sulfur wth 65 wt. % JSC-1 sample tested at 21 °C (70 °F) .................................................................. 29 40. SEM mcrograph of a fracture surface from cycled sample 20 wth the same composition as figure 39, but tested at –101 °C (–150 °F) ................................................. 30 41. Photograph of a sulfur sample wth 65 wt. % pure SO2 ................................................... 32 42. Photograph of a sample like that seen in figure 7 which was cycled 20 times between RT and –196 °C (–321 °F). Crumblng of the sample wth free grans of slca s seen ................................................................................................................... 32 v

Original page 9

LIST OF ACRONYMS AND SYMBOLS °C degrees Celsus °F degrees Fahrenhet °R degrees Rankne cm centmeter FeS troilite, iron sulfide g gram JSC-1 Johnson Space Center lunar sol smulant K Kelvn degree LN2 lqud ntrogen m meter mg mllgram mm mllmeter MPa megapascal Pa Pascal ps pounds per square nch S sulfur Sb antmony SEM Scannng Electron Mcroscope SO2 slca, slcon doxde wt. % percent by weght v

Original page 10

v

Original page 11

NOMENCLATURE L orgnal length m atomc mass P ambent pressure P* partal pressure of gas n equlbrum wth ts sold R gas constant RT room temperature T temperature vf volume fracton α coefficient of thermal expansion αv evaporation coefficient Γ mass evaporaton rate ∆L change n orgnal length ∆T temperature dfference ε stran x

Original page 12

x

Original page 13

TECHNICAL MEMORANDUM SULFUR ‘CONCRETE’ FOR LUNAR APPLICATIONS— ENVIRONMENTAL CONSIDERATIONS 1. INTRODUCTION Conventonal concrete conssts of sand, a coarser aggregate, and a hydraulc bnder based on calcum slcate. Added water chemcally reacts wth the calcum slcate whch then effectvely sets up and hardens nto the mass known as concrete. Sulfur ‘concrete’ s somewhat a msnomer, as very lttle, f any, chemcal reacton occurs between the consttuent materals. Bascally the sulfur, a thermoplastc materal, s melted and mxed wth an aggregate, after whch the mxture s poured, molded, and allowed on materal1–17that has ganed wde acceptance, parto harden. As such, t s an establshed construct ticularly for use in environments subjected to acids and salts, figures 1–4.1It exhbts good compressve strength (generally better than Portland cement), low water permeablty, and rapd setup tmes. Fgure 1. Eght-cubc-yard heater-mxer moble producton unt. The composte composton generally ranges from 12–22 percent by weght, or weght percent (wt. %) sulfur and 78–88 wt. % aggregate, whch can consst of any number of materals, ncludng rock sands, minerals, fly ash, rubber particles, and glasses. The mixture can also contain some 5% of a group of compounds termed plasticizers that mitigate cracking as the sulfur goes through, at ≈96 °C (≈205 °F), a reversble monoclnc-rhombc crystallne phase change. One downsde s sulfur’s narrow workng range. It melts at ≈120 °C (248 °F), but above 148 °C (298 °F) the lqud experences a phase change where it ‘stiffens’ and loses needed fluidity. Making and applying sulfur concrete is generally constraned between 130 and 140 °C (266 and 284 °F) and, obvously, t cannot be used n an envronment that exceeds ≈120 °C (248 °F). 1

Original page 14

Fgure 2. Pourng of Chempruf Sulfur Concrete n an ndustral plant. Fgure 3. Chempruf Sulfur Concrete: acd contanment system. Fgure 4. Precast Chempruf Sulfur Concrete tanks for Aqua Rega. n the form of the mneral trolte, FeS,18,19whch rases the Sulfur has been found on the Moon  queston of reducng the ore to obtan sulfur for constructon purposes. Ths s an attractve alternatve to conventonal concrete, as water, a precous resource, s not requred. Reducng trolte to elemental 2

Original page 15

ng sulfur concrete on the Moon has been prevously dscussed.20–27For our purposes t sulfur and us s assumed that elemental sulfur s avalable on the lunar surface and sulfur concrete products such as brcks can be made. Acknowledgng that envronmental condtons on Earth are relevant to the use of sulfur concrete, t follows that lunar applcatons would ental addtonal concerns. Lunar temperatures at the equator range from +123 to –180 °C (253 to –292 °F), wth an average of –20°C (–4 °F) and, at the poles, from –60 to –220 °C(–76 to –364 °F). These are extreme temperatures and, perhaps more mportant, extreme temperature cycles. How ths mght affect the mechancal propertes of sulfur concrete s unknown. Secondly, the Moon’s envronment s also characterzed by a lack of atmosphere, generally assumed to be on the order of 1 × 10–12torr (1.33 × 10–10Pa). Ths low pressure brngs nto queston sublmaton processes where a materal that s a stable sold at 1 atmosphere (atm) would now transform to a gaseous state. Clearly t can be expected that these extremes n temperature and pressure wll affect the vablty of sulfur concrete. The followng sectons wll evaluate, as best as possble, these envronmental concerns, wth the ntent of ascertanng the feasblty of usng sulfur concrete as a constructon materal on the lunar surface. 3

Original page 16

1. SUBLIMATION CONSIDERATIONS A well known example of sublmaton on Earth s where sold carbon doxde (dry ce), upon warmng, drectly transforms to gas. Some nsght to sulfur sublmaton can be ganed by examnaton of ts unary pressure-temperature phase dagram.28Note that ths phase dagram s not partcularly well understood and several generally similar versions, such as the one seen in figure 5, can be found. 104 Rhombic 102 1 Monoclinic 10–2 Pressure (atm) 10–4 96 oC 10–6 50 100 Temperature (oC) 151 oC Liquid 120 oC Vapor 150 200 Fgure 5. Typcal unary sulfur phase dagram. Gven that, on Earth at 760 torr (1 atmosphere, 0.1 MPa), as the temperature rses from room temperature (≈20 °C (68 °F)), sulfur undergoes a solid phase transition—here ≈96 °C (205 °F)—and F). In contrast, at pressures on the order of 7.6 × 10–3torr (1.0132 Pa), then melts at ≈120 °C (≈248 ° solid sulfur transforms directly, or sublimates, to a gaseous phase at ≈90 °C (194 °F); the transton ng pressure.28, 29Recall the pressure on the Moon, temperature contnues to decrease wth decreas ≈1 × 10–12torr (1.33 × 10–10Pa), and t s realstc to assume that the sulfur composng sulfur concrete 4

Original page 17

would be prone to sublmate, resultng n a deterorated, unsound, structure. Wth that n mnd, the ntent of ths work s to study the effect of a hgh and prolonged vacuum on pure sulfur and two presumed lunar-lke sulfur ‘concrete’ compostons. 2.1 Experimental Procedure Pure sulfur and sulfur ‘concrete’ mxtures of two dfferent compostons were prepared for evaluation. The pure sulfur was melted and cast into small, circular, hard plastic molds ≈45 mm (1.8 in) n dameter and 5 mm (0.2 n) deep. The approxmate compostons of the concrete samples were: th 65 wt. % JSC-1 (an establshed lunar smulant sol30that has a sand-lke (1) 35 wt. % pure sulfur w consstency) and (2) a sulfur and slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1. The sulfur-slca bnder s a commercally avalable product known as Glson Redron 9000 Cappng Compound, Amercan Socety for Testng Materals (ASTM) C617 and Amercan Association of State Highway and Transportation Officials (AASHTO) T231. The sulfur, or binder mixture, s melted and mxed wth the aggregate whch s then poured nto heated molds havng dmensons of 2 × 2 × 2 in (≈50 mm each side) with excess added to account for shrinkage. The cube is allowed to cool, after whch t s removed and readed for testng, a procedure that generally entals sectonng the block into pieces such as seen in figure 6. Additional experimental details can be found elsewhere.25–27 Fgure 6. A pece of sulfur-based concrete made at The Unversty of Alabama n Huntsvlle. The corners were broken off for testng purposes. The sublimation experiments were conducted in a vacuum chamber, figure 7, which is capable of achevng a vacuum level on the order of 5 × 107torr (6.67 × 10 Pa), a level below whch sublmaton 5 of sulfur at 20 °C (68 °F) is expected upon extrapolation of the pressure/temperature lines shown in figure 8, another varaton of the unary sulfur pressure-temperature dagram. Here the green square marks the prescrbed expermental condtons and suggests that sublmaton of the sulfur from the concrete composte wll occur. 5

Original page 18

Fgure 7. Photograph of the vacuum chamber used for the sublmaton experments. To verfy that sublmaton under the proposed expermental condtons—room temperature (RT) and 6.67 × 10–5 Pa—would actually occur, one of the broken-off corners of the sample shown in figure 6 was placed in the chamber. After approximately 6 days at ≈3 × 10–6 torr and ≈20 °C (≈68 °F), the sample was removed for examnaton. Fgure 9 shows the pece subjected to vacuum processng, n the upper rght hand corner, matched wth the untested secton from whch t was broken. The surface of the processed sample exhbts obvous porosty enhancement and a granular morphology. It s further noted that the ntal pece weghed 2.1449 g (0.0047 lb) and after removal weghed 2.0416 g (0.0045 lb), essentially losing ≈0.1g (0.0002 lb) of sulfur. The loss is not likely due to water vapor or other volatiles, as the sample was placed n an oven overnght at 110 °C (230 °F) pror to ts beng weghed and placed n the vacuum chamber. For the controlled sublmaton study, small sectons on the order of 10 mm × 5 mm × 3 mm (0.393 × 0.196 × 0.116 n) were cut from a cube representatve of the two concrete compostons and weghed. One pece of each composton was placed, large surface area down, n a small alumnum weghng dsh; sx sets were prepared. The sx sample sets were then placed n the chamber and subjected to vacuum processng. Every 5 to 10 days, the chamber was opened and the samples removed for weghng; one composton set was kept out and the remanng were put back n the vacuum chamng, a vacuum level on the order of 3–7 × 10–6torr (4–9.3 × 10–4Pa) was observed ber. Generally speak durng processng. The samples removed durng testng were weghed and ther surfaces photographed. The exposed surface area of each sample was measured and the recorded weght loss from sublmated meter (mg/mm2). Ths scenaro was kept up, more sulfur was expressed as mllgrams per square mll or less, for 60 days. A smlar scenaro was ndependently conducted for the pure sulfur samples. The pure sulfur samples were kept n the plastc molds descrbed above, ther surfaces left to be exposed to the vacuum. 6

Original page 19

104 103 102 101 (torr) 1 760 10–1 76 Pressure (atm)–2 10 7.6 10–3 0.76 Rhombic Liquid Monoclinic 10 7.6s10–2 –4 10–5 0 10 20 30 40 50 60 70 7.6s10–5 Temperature (oC) 7.6s10–6 7.6s10–7 Vapor 90 100 110 120 130 140 150 160 Fgure 8. Extrapolaton of the sold-vapor transton lne on a unary sulfur phase dagram. Fgure 9. Comparson of the sulfur concrete surfaces after vacuum processng of the upper rght-hand corner secton. 7

Original page 20

2.2 Experimental Results Fgures 10–12 show representatve surfaces of the as-cast pure sulfur and samples that have been exposed to vacuum. First of all, the surface of an ‘as-cast’ sample, figure 10, is not smooth or uniform, the obvious irregularities being shrinkage effects such as cavities and dendrite in-filling with a glazed appearance. After 11 days in vacuum, as shown in figure 11, sulfur sublimation clearly reveals the intricate nature of the prmary dendrtes. Fgure 12 shows a typcal surface after 54 days n vacuum. Here the shrinkage cavities have grown and the fibrous nature of the dendritic crystals is accentuated. Fgure 10. Surface morphology of the as-cast pure sulfur. Fgure 11. Surface morphology of the as-cast pure sulfur after vacuum processng on the order of 3–7 × 10–6torr (4–9.3 × 10 Pa) for ≈11 days. –4 8

Original page 21

Fgure 12. Surface morphology of the as-cast pure sulfur after vacuum processng on the order of 3–7 × 10–6torr (4–9.3 × 10 Pa) for ≈54 days. –4 Fgures 13 and 14 show, respectvely, the surfaces of the samples removed from vacuum after 8, 15, 25, 39, 46, and 58 days. Samples in figure 13 have the sulfur - 65 wt. % JSC-1 (Johnson Space Center lunar soil simulant) composition, and those in figure 14 consist of the sulfur and silica binder mixture Fgure 13. Sulfur - 65 wt. % JSC-1 samples subjected to vacuum processng. Numbers ndcate days processed. 9

Original page 22

Fgure 14. Sulfur and slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 subjected to vacuum processng. Numbers ndcate days processed. (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1. Obvous and ncreasng degradaton of the samples as a function of time is seen, particularly in figure 13, whose samples have 10% more sulfur than those in figure 14. Fgures 15–20 are representatve mcrographs of sample sulfur ‘concrete’ surfaces shown above. Figure 15 shows the surface of the as-cast, sulfur - 65 wt. % JSC-1, sample, figure 16 shows that sample after 8 days’ exposure to vacuum, and figure 17 after 58 days. Figures 18–20 are similar except that the samples contan SO2and less sulfur. The as-cast surfaces are relatvely smooth; clear degradaton of the surface s seen after 8 days, more so after 58 days. Agan, the extent of degradaton s obvously more apparent in figures 17 and 18, showing the concrete samples containing the greatest amount of sulfur. Fgure 21 plots the measured sulfur weght loss, normalzed to mllgram per square mllmeter (mg/mm2) of exposed surface area, as a functon of tme. As seen, the number of ponts for a gven composton decreases wth tme due to the removal of one sample at each examnaton perod to assess surface degradaton. Although scatter wthn the ndvdual samples s seen, clear trends for the gven sample compositions are apparent. The graph shows, as figures 10–12 and 15–20 suggest, the greater the amount of ntal sulfur n the sample, the more that wll sublmate away over a gven perod. The rate of sublmaton also decreases wth tme, and one mght also nfer, partcularly for the aggregate contanng samples, that they wll reach some constant value. 10

Original page 23

Fgure 15. Mcrograph showng the surface of the as-cast sulfur - 65 wt. % JSC-1 sample. Fgure 16. Mcrograph showng the surface of the sulfur - 65 wt. % JSC-1 sample after 8 days n vacuum. 11

Original page 24

Fgure 17. Mcrograph showng the surface of the sulfur - 65 wt. % JSC-1 sample after 58 days n vacuum. Fgure 18. Mcrograph showng the surface of the as-cast sulfur and slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 sample. 12

Original page 25

Fgure 19. Mcrograph showng the surface of the slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 sample after 8 days n vacuum. The sphercal partcles are the SO2grans. Fgure 20. Mcrograph showng the surface of the slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 sample after 58 days n vacuum. 13

Original page 26

1 Pure Sulfur 0.9 Sulfur z65 wt.% JSC-1 0.8 Sulfur z55 wt.% JSC-1 z20 wt.% SiO2 Temperature = 18 to 20 oC ) 0.7 2 0.6 0.5 0.4 Weight Loss (mg/mm 0.3 0.2 0.1 0 0 5 10 15 20 25 30 35 40 45 50 55 60 Time (days) ght loss for the exposed surface area (mg/mm2) Fgure 21. Plot of the sulfur we as a functon of tme. 2.3 Discussion of Sublimation Concerns Fgure 21 shows clear trends n sulfur sublmaton as a functon of sample materal. Stll, there s obvous scatter n the weght loss/surface area for a gven set of samples. Ths s attrbuted to several factors. For nstance, the actual surface area measurement of the sample s probably off by ±5%. In addition, the surface area measurement is strictly a bulk sample consideration. Examination of figures 10, 15, and 18 clearly shows that the stated surface area does not consder surface rregulartes such as ledges and porosty. After vacuum processng, the sublmated vods further add to the surface area. Addtonal scatter n the data could arse from nonunform dstrbuton of aggregate; more at the surface on a gven sample would show less weght loss. It s also assumed that the sublmaton rate s unform on the sample surface and sdes; the bottom area n contact wth the alumnum pan s gnored. Some insight to this latter assumption was inadvertently obtained. In figure 21 the green diamond datum point representng the sulfur - 65 wt. % JSC-1 sample after 8 days s uncharacterstcally hgh. Examnaton of the sample showed that after cuttng the sample from the bulk, a small ‘leg’ was unntentonally left on the sample,26 as shown in figure 22. Ths effectvely rased the sample bottom from the alumnum pan and exposed addtonal area from whch sulfur could easly sublmate. One must also consder that over tme, as the sulfur sublmates away, more and more of the aggregate s exposed. In ths case, any now ‘free’ aggregate would st on the sample surface, effectvely blockng any sulfur underneath from leavng. If ths same pece s placed on ts sde, ts large surface area now parallel to a gravty vector, exposed aggregate may well fall off, exposng fresh sulfur for sublmaton. In summary, scatter n the observed data should be expected and s dependent on, at least, aggregate sze, shape, dstrbuton, and sample exposure poston. 14

Original page 27

Fgure 22. Macrograph of the sulfur - 65 wt. % JSC-1 sample after 8 days n vacuum that represents the green diamond datum point, figure 21. The ‘leg’ inadvertently left on the sample effectvely exposed more surface area and resulted n a greater mass of lost sulfur. Consder now the sublmaton data for the ndvdual sample compostons: (1) pure sulfur, (2) sulfur wth 65 wt. % JSC-1, and (3) the sulfur-slca bnder and 55 wt. % JSC-1 mxture. Obvously, as seen, the sample wth the most sulfur per unt bulk surface area wll sublmate away the most. Gven that, weght loss per surface area for pure sulfur appears nearly constant wth tme, whereas the other two compostons gve the mpresson of decreasng slghtly wth tme. The evaporaton rate of sulfur ng the well known Hertz-Knudsen equaton31–33gven below. can be evaluated us  m Γ = αν   2πRT 1 2/   (P * –P) . (1)  Here Γ s the mass evaporaton rate, αv is designated as an evaporation coefficient, m s the s the gas constant (8.31432 J⋅K–1⋅mol–1), T s the absolute temperaatomc mass (32.06 for sulfur), R  ture (293 K for ths work), P* s the partal pressure of the gas phase n equlbrum wth ts sold (assumed to be 1.8 × 10–6torr or 2.4 × 10–4Pa), and P s the ambent (obtanable) pressure of the expermental vacuum chamber (5 × 10–7torr or 6.67 × 10–5 Pa). The evaporation coefficient, α , was ntroduced v and justified by Knudsen as a consequence of experimental results being less than that predicted by equation (1) in its basic form. Also known as a ‘sticking’ coefficient, αv is a measure of the difficulty for atoms to ether attach to or be released from a surface, and ascertanng ts value s fraught wth dfficulty.34Values of α ≈ 1 were reported for tungsten, copper and ron, nckel and nckel oxde, v 35 36 37 and beryllum.38Other values for α nclude 0.17 for Sb at 650 K (1,170 °R), 0.17 for LF at 1,000 K v 39 (1,800 °R),40and 0.1–0.4 for the KCl-NaCl system between 913 K (1,643 °R) and 1,033 K (1,859 °R). 41 A value of 4.6 × 10–3was also measured for arsenc at 550 K (990 °R). No value of α for sulfur could 42 v be found, so t was, for purposes of calculaton, assumed to equal 1. 15

Original page 28

Usng the above values, evaluaton of equaton (1) results n an evaporaton rate for pure sulfur of Γ = 2.50808 × 10–8g⋅cm–2⋅s–1(2.50808 × 10–7mg⋅mm ⋅s ); ths deally translates, at 20 °C (68 °F), –2 –1 to sublimating away a 1-cm-thick (0.4-in-thick) layer of sulfur in ≈955 days. Consequently, assuming a unform sample surface area of 962.1 mm2, one might expect to lose ≈2.413 × 10 mg/s of sulfur from –4 the sample. If ths value remaned constant for the length of the experment (60 days at 18–20 °C, or 64–68 °F), 1.251 g of sulfur would sublmate. Expermentally, the pure sulfur sample that was exposed to vacuum for 60 days lost 0.9329 g. The Hertz-Knudsen equation for pure sulfur and the experimental conditions are plotted in figure 23. Examnaton reveals good agreement, assumng αv = 1, for the first 10 days or so, after which ncreasng devaton s seen. 1 Pure Sulfur 0.9 Sulfur z65 wt.% JSC-1 Sulfur z55 wt.% JSC-1 z20 wt.% SiO2 0.8 H-K Theory (Av=1) 0.7 Temperature = 18 to 20 oC ) 2 0.6 0.5 0.4 Weight Loss (mg/mm 0.3 0.2 0.1 0 0 5 10 15 20 25 30 35 40 45 50 55 60 Time (days) Fgure 23. Comparson of measured weght loss for pure sulfur wth the Hertz-Knudsen equaton, αv = 1. The higher initial rate of sulfur sublimation can be inferred from figures 10–12. This as-cast material has the greatest surface area, as shown in figure 10, from which to draw atoms. The surface is essentially composed of a dendritic network, the arms in-filled with the last material to solidify, giving it a glazed appearance. This in-filled surface material would contain the majority of any impurities, have a slghtly lower meltng temperature, and possbly have greater volatlty. Ths postulate s supported by figure 11, which shows emergence and clear definition of the primary (highest melting point) dendritic structure. The surface seen in figure 11 has also been subjected to vacuum for 11 days, approximately the time at which the sublimation rate is seen to begin decreasing, (see fig. 23). This decrease in 16

Original page 29

sublmaton rate can be attrbuted to at least two factors. Frst, the cavtes would stll contan the resdual solidification product, but their deepening now increases the difficulty for sulfur atoms to leave. Second, the surface now consists of well exposed, but randomly oriented, dendrites that grow in specific crystallographc drectons. The exposed dendrte arms consttute a mesh that could obstruct released s further expected that the release rate of atoms wll dffer (e.g., Inaba43) for each crystalatoms, and t  lographc orentaton. Thus t s prudent to note that the assumpton of αv = 1 may have been fortutous and that an appropriate evaluation of its true value would entail specifically designed equipment in conjunction with a sulfur single crystal of known orientation. Finally, figure 12 (54 days in vacuum) shows ncreasng sze of the shrnkage cavtes and further detal of the dendrte structure; the evaporaton rate contnues to slowly decrease. Obvously, as the aggregate mxtures contaned less sulfur, sublmaton rates were expectedly ) of sulfur and SO are,44respectvely, 2.07 g cm–3and less for them, figures 16–20. The densities (ρ 2 2.32 g cm–3; the densty30of JSC-1 s 2.9 g cm–3. Converting the sulfur ≈65 wt. % JSC-1 composition and 25% sulfur and 20% slca by weght wth 55 wt. % JSC-1 composton samples to volume percentages finds the former to be 57.0 volume percent aggregate (43% S) and the latter to be 69.55 volume percent aggregate (30.45% S). To a first approximation, reasonably noting that the aggregate material does not sublmate, volume fracton factors, vf, (0.43 and 0.3) can be placed n conjuncton wth the evaporation coefficient; the Hertz-Knudsen equation with this correction factor included is plotted for the two mixtures in figure 24. Again, early fit with the data is seen, but the theory soon over-predicts the experimental results. This is expected, particularly in view of figures 15–20. Initially the sample surfaces are relatively smooth wth lttle, f any, of the aggregate exposed. More and more of the aggregate s exposed as sublmaton proceeds. Ths effectvely reduces the amount of exposed sulfur, assumng the aggregate does not fall away, which over time leaves less to sublimate through an increasingly difficult path. Figure 25 shows the surface of two samples that were exposed to vacuum for 60 days. On the left s the sulfur - 65 wt. % JSC-1 sample and on the right is sulfur ≈55 wt. % JSC-1 ≈20 wt. % SiO2. The surface dscoloraton s a consequence of mechancally removng the vacuum affected materal. Although the sulfur sublimed, the aggregate particles remained cohesively in place, e.g. figures 13 and 14. However, they could be easly removed by a dsturbance as mld as gentle tappng. The vacuum-exposed surface of the sulfur - 65 wt. % JSC-1 sample (left, in fig. 25) is the reddsh-brown trangular shape located at the central rght. The darker brown regon surroundng t s the materal underneath and was easly exposed by a small ar jet. The lghter strpe along the left sde denotes the regon where the brown materal was gently scraped away, exposng hard and ntact material. The exposed surface of the sulfur ≈55 wt. % JSC-1 ≈20 wt. % SiO2 sample (fig. 25, right) is the wder strp of brownsh-orange on the rght sde. The lghter strp on the left s where the loosely cohesve materal was gently scraped away. The surface area of the samples s approxmately 6 mm × 11 mm. Over a period of 60 days, see figure 24, the samples should have, left and right, lost 0.02178 and 0.01056 g of sulfur. For a surface area of 66 mm2, these values translate to sulfur thcknesses of 0.159 mm and 0.0773 mm. Recall that the sulfur only accounts for ≈43.0 and ≈30.45 volume percent in these samples. When the aggregate s ncluded, the overall volume fractons of the affected sample materal are, respectvely, 24.47 mm3and 16.75 mm3. Ths translates to thcknesses of 0.37 mm and 0.254 mm. In summary, the calculated thcknesses of the vacuum-affected zones are reasonably representatve of 17

Original page 30

1 Pure Sulfur 0.9 Sulfur z65 wt.% JSC-1 Sulfur z55 wt.% JSC-1 z20 wt.% SiO2 0.8 H-K Theory (Av=1) 0.7 H-K Theory, 43 vol. % S. (Av=1, vf =0.43) ) 2 0.6 H-K Theory, 30.45 vol. % S. (Av=1, vf =0.3) 0.5 Temperature = 18 to 20 oC Weight Loss (mg/m0.4 0.3 0.2 0.1 0 0 5 10 15 20 25 30 35 40 45 50 55 60 Time (days) Fgure 24. Comparson of the measured weght loss for pure sulfur and the sulfur concrete samples wth the Hertz-Knudsen equaton, αv = 1, and wth volume fracton, vf, correcton factors. what was experimentally seen. What has been verified, at least for these samples and conditions, is that the affected depth s a functon of the aggregate volume fracton. Put another way, assumng that the sulfur loss rate for the concrete samples becomes constant as suggested by figure 6, it would take ≈4.4 yr to sublimate to a depth of 1 cm in a sulfur ≈65 wt. % JSC-1 brick versus ≈6.5 yr for the sulfur ≈55 wt. % JSC-1 ≈20 wt. % SiO2composton. Ths, agan, s drawn from data where the bulk of the exposed aggregate remaned on the sample surface. The experimental results and analysis presented above were confined to room temperature, i.e., ≈20°C (68 °F). Lunar temperatures vary consderably from ths (–230 to 130 °C, or –382 to 266 °F) and can significantly affect sublimation rates. Utilizing the Hertz-Knudsen equation in conjunction with the vapor pressure versus temperature curve,29 the calculated curves shown in figure 26 plot the time to sublmate away a 1-cm-thck sulfur layer. Two pressures (Moon and expermental chamber) are consdered between temperatures rangng from 15 to 120 °C (59 to 248 °F), the latter beng near sulfur’s meltng point. At ≈15 °C (≈59 °F), the effect of a much lower lunar pressure s seen and a 1-cm (0.4-n) layer s calculated to take 3.7 yr to evaporate n comparson to 8.4 yr, whch s calculated for the ground-based experimental chamber. Above ≈30 °C (86 °F) , the two curves are essentally ndstngushable. However, the consequence of ncreasng temperature becomes obvous. What mght be tolerable at 15 °C 18

Original page 31

Fgure 25. Photograph of two samples exposed to vacuum for 60 days: left: sulfur - 35 wt. % JSC-1 and rght: 25% sulfur and 20% slca by weght wth 55 wt. % JSC-1. 120 100 120 oC, 1.63 hr 80 C) o 60 Temperature ( 40 20 0 0.01 0.1 1 P = 1s10–12torr (Moon) P = 5s10–7torr (experiment) 15 oC, 8.4 yr 15 oC, 3.7 yr 10 100 1,000 10,000 Days to Sublimate a 1-cm Sulfur Layer Fgure 26. Calculated plot showng the effect of temperature on the tme needed to sublmate a 1-cm (0.4-n) thckness of pure sulfur. 19

Original page 32

(59°F) is clearly not at ≈120 °C (248 °F), where t s calculated to take less than 2 hr to sublmate a 1-cm (0.4-n) layer. The above work consders vacuum effects over tme on pure sulfur and two sulfur concrete compositions at a temperature of ≈20 °C (68 °F). Other temperatures, or even cycles, can easly be mplemented on the system to mimic lunar conditions and ascertain evaporation rates. What is difficult to s on the order of 10–12torr (1 × 10–10Pa). Ths rases the queston reproduce s the lunar pressure, whch  of whether or not the results presented here, for the gven temperature, are applcable to a lunar envronment. Recall that the partal pressure of sulfur (P*) n equlbrum wth ts sold (at 20 °C, or 68 °F) was assumed to be 1.8 × 10–6torr (2.4 × 10–4Pa) and that the vacuum chamber was able to reach an ambent level (P) of ≈5 × 10–7torr (6.67 × 10–5Pa). For a constant P* and T the evaporaton rate (Γ ), equaton (1), should ncrease as P decreases. Thus, s t reasonable to assume that a laboratory chamber capable of 5 × 10–7torr s representatve of the lunar envronment, whch s smaller by some 5 orders of magntude? The calculated results shown in figure 27 gauge the role vacuum level plays on the sublimation rate by ronmental (ambent) pressure. At P = 1.8 × 10–6torr, (P*–P) s plottng (P*–P) as a functon of the env zero, but as P decreases, the desgnated degree of sublmaton rapdly ncreases, beng upper bound by a maxmum of 1.8 × 10–6when P equals zero. For an envronmental pressure of 5 × 10 torr, ths cor- –7 s 72.2% of the value achevable. Decreasng the pressure to 1 × 10–7 responds to a (P*–P) value that  torr accounts for 94.4% and 1 × 10–8torr relates to 99.4%. Whle t mght be prudent to ncrease the measured sublimation rates by ≈28%, it is apparent that the much lower lunar atmosphere of 10–12torr would contrbute only neglgbly. 2s10–6 Degree of Sublimation 1.8s10–6 1.6s10–6 1.4s10–6 1.2s10–6 1s10–6 8s10–7 6s10–7 Equilibrium Vapor Pressure (P*–P) 4s10–7 2s10–7 0 1s10–5 1s10–6 1s10–7 1s10–8 1s10–9 Ambient Pressure (P, torr) Fgure 27. Calculated plot showng the role the ambent pressure plays n regard to the equlbrum vapor pressure. 20

Original page 33

2.4 Summary of Sublimation Concerns Pure sulfur and two sulfur concrete mxtures were prepared and placed n a vacuum envronment (capable of 5 × 10–7 torr) at ≈20 °C (68 °F) for 60 days. Perodc weghng of the samples revealed a contnuous weght loss caused by the sublmaton of sulfur. The sublmaton rate was evaluated wth the Hertz-Knudsen equation, assuming an evaporation coefficient of 1.0. Reasonable agreement over ≈10 days was found for pure sulfur and for the concrete mixtures when the volume fraction of added aggregate was consdered. Subsequent dscrepances were attrbuted to nonunform surfaces, mpurtes, and contnual exposure of aggregate materal. The dfference n volume fracton of aggregate between the two concrete samples (57% and 69.6%) was reflected in the depth of affected (sublimated) material. Here, for the gven condtons, t was predcted that 4.4 and 6.5 yr, respectvely, would be needed to sublmate away a 1-cm-deep (0.4-n-deep) layer from the concrete samples. Sulfur sublmaton rates were predcted to change dramatcally over a temperature range from 15 to 120 °C (59 to 248 °F). Fnally t was shown that the much lower vacuum on the Moon would contrbute only slghtly more to the sublmaton rates determned from the ground-based experments. 21

Original page 34

1. EXTREME COLD AND TEMPERATURE CYCLE CONSIDERATIONS REGARDING THE INTEGRITY AND COMPRESSION STRENGTH OF SULFUR CONCRETE The prevous secton addressed the effect of sulfur sublmaton n the hard vacuum that exsts on the lunar surface and the consequence t could have f sulfur concrete were to be used as a constructon materal. It s obvous that sulfur concrete cannot be used n an envronment where the temperature exceeds ≈120 °C (248 °F)—sulfur’s meltng temperature. It has also been shown that the sublmaton rates below ths temperature are relatvely hgh and that a sulfur concrete product would quckly lose ts ntegrty. On the other hand, at the lower lunar temperatures, e.g., –180 to –220 °C (–292 to –364 °F), the sublmaton knetcs are so slow as to be essentally neglgble. However, one must also consder the effects of such low temperatures and extreme temperature cycles on the mechancal propertes of sulfur concrete, partcularly as t s a composte materal. The followng subsectons descrbe an effort to gan some nsght regardng the mechancal propertes of sulfur concrete that mght be subjected to the extreme temperatures of the lunar envronment. 3.1 Experimental Procedure The expermental procedure, for the most part, was descrbed n secton 2. Agan, the same two compositions were employed: (1) Thirty-five wt. % pure sulfur with 65 wt. % JSC-1, and (2) a sulfur and slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1. The molten sulfur mxtures were cast nto 50.8-mm3(2-n3) blocks and allowed to harden. A gven block was then cut nto eight 25.4-mm (1-in) cubes such as seen in figure 28. Blocks that showed significant imperfections such as large shrnkage cavtes were dscarded. Fgure 28. Typcal 2.54-cm (1-n) cube of sulfur concrete used n ths study. Sample composton s a slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1. 22

Original page 35

A set of eght cubes, four of each composton, was packaged wth an enclosed k-type thermocouple and placed in the bottom of a wide mouth flexible plastic foam container into which liquid ntrogen (LN2) was poured. The contaner was capped wth foam rubber, allowng the LN2to cool the samples and then evaporate over time. This cycle was repeated 80 times between RT (≈20 °C, or 68 °F) and LN2 (≈ –191 °C, or –312 °F) temperatures. A typical time-temperature plot is shown in figure 29. 50 0 C) o –50 –100 Temperature ( –150 –200 –250 0 200 400 600 800 1,000 1,200 Time (min) Fgure 29. A typcal tme temperature plot showng one of the cycles the samples experenced. Cycled and noncycled samples of both compostons were then subjected to compresson testng at a constant downward crosshead speed of 0.127 cm/mn (0.05 n/mn). One set of samples was tested at room temperature (≈21 °C, or 70 °F) and the other at ≈–101°C (≈ –150 °F), figure 30. Compression data were gathered and the tests stopped after obvous crushng was observed. 3.2 Experimental Results The noncycled samples were typical of that seen in figure 1, whereas those subjected to 80 cycles between RT and LN2 temperatures exhibited cracking on the surface, such as seen in figure 31. The maximum strength achieved during compression testing of a given sample is shown in figure 32. Samples 1–16 were noncycled, whereas 17–24 were cycled. Informaton regardng sample composton and test temperature is located in the figure legend. 3.3 Discussion of Extreme Temperature Concerns Consider first the noncycled samples, numbers 1–16. Samples 1–8 represent those consisting of sulfur - 65 wt. % JSC-1. The first four were tested at 21 °C (70 °F) and the latter four at –101 °C (–150 °F). Samples 9–16 represent the sulfur - 55 wt. % JSC-1 and 20 wt. % slca samples. Agan, the first four were tested at 21 °C (70 °F) and the latter four at –101 °C (–150 °F). Lookng at the group as a whole, the maximum compression ranges from ≈17 MPa to ≈47 MPa. In short, there are statistically insufficient data to make any conclusions regarding differences based on composition and/or test 23

Original page 36

Fgure 30. Photograph of a sample undergong compresson testng at –101 °C (–150 °F). Fgure 31. Photograph of crackng exhbted on the surface of sample 22, which was subjected to 80 cycles between RT (≈21 °C, or 70 °F) and –191 °C (–312 °F). 24

Original page 37

50 Sulfur Concrete Compression Tests 45 40 35 30 25 20 Compression (MPa) 15 10 5 0 Noncycled Samples S-65 wt. % JSC-1, 21 oC S-65 wt. % JSC-1, –101 oC S-55 wt. % JSC-1, 20 wt. % SiO2, 21 oC S-55 wt. % JSC-1, 20 wt. % SiO2, –101 oC Cycled Samples S-65 wt. % JSC-1, 21 oC S-65 wt. % JSC-1, –101 oC S-55 wt. % JSC-1, 20 wt. % SiO2, 21 oC S-55 wt. % JSC-1, 20 wt. % SiO2, –101 oC 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Sample Number Fgure 32. Plot of the maxmum compresson strength exhbted by the cycled and noncycled samples. temperature. The dscrepancy n results shown here s lkely a consequence of defects such as nhomogeneous aggregate dstrbuton and/or porosty that would arse durng freezng of the concrete mxture. The cycled samples, numbers 17–24, follow the same scenaro as above, except that two nstead of four samples per set were used. Here the test data range from ≈3 MPa to ≈8 MPa. Although the silicacontanng samples appear to have performed slghtly better, as one mght ntutvely thnk, there are no statstcs to back that observaton. In summary, ndvdual dfferences due to temperature or composton cannot be ascertaned for ether the cycled or noncycled sample groups. However, there s a clear dfference between the maxmum compresson strength obtaned from the cycled samples when compared to those noncycled samples. Assuming average compression failures of ≈35 MPa for the noncycled and ≈7 MPa for the cycled samples, a dfference of about 5 tmes s seen, a factor that s easly attrbuted to the cracks observed on the cycled samples. Ths premse was nvestgated by examnng the fracture surfaces and s best exemplified in the silica-containing samples. Fgures 33–36 are representatve scannng electron mcrographs of fracture surfaces from samples noted in figure 32. Figure 33 is taken from sample 10, a noncycled silica binder (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1 mxture tested at 21 °C (70 °F). Fgure 34 s from sample 14, which is like sample 10 in figure 33, but tested at –101 °C (–150 °F). Fgure 35 s from sample 22, a cycled slca bnder (25% sulfur and 20% slca by weght), wth 55 wt. % JSC-1 mxture tested at 21 °C (70 °F). Figure 36 is from sample 24, which is like sample 22 in figure 35, but tested at –101 °C (–150 °F). 25

Original page 38

Fgure 33. SEM mcrograph of a fracture surface from noncycled sample 10, a slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1, tested at 21 °C (70 °F). Fgure 34. SEM mcrograph of a fracture surface from noncycled sample 14, with the same composition as in figure 33, but tested at –101 °C (–150 °F). 26

Original page 39

Fgure 35. SEM mcrograph of a fracture surface from cycled sample 22, a slca bnder mxture (25% sulfur and 20% slca by weght) wth 55 wt. % JSC-1, tested at 21 °C (70 °F). Fgure 36. SEM mcrograph of a fracture surface from cycled sample 24, wth the same composition as in figure 35, but tested at –101 °C (–150 °F). 27

Original page 40

Fgure 33 s from noncycled sample 10 and shows a mottled fracture path through the sulfur and around the sphercal slca partcles. Note that the partcles have mantaned good coherency wth the sulfur bnder, some of whch can be seen adherng to the surface of the large partcle at the top center of the mcrograph. Sample 10 also had one of the lowest compresson strengths. Ths s lkely due to porosty, as can be seen n the upper rght hand corner, whch was n evdence throughout ths secton. Sample 14 was tested at –101 °C (–150 °F) and figure 34 shows the fracture surface morphology. Bonding of the silica particles with the sulfur is again evident, and the surface looks like that of figure 33, less the porosty; any dstncton due to the lower test temperature s not obvous. Fgure 35 s from cycled sample 22, tested at 21 °C (70 °F). Here, in contrast to figure 33, spherical silica particles are seen lying on the surface fully de-bonded from the sulfur binder. This is again seen in sample 24, figure 36. Fgures 37– 40 are representatve scannng electron mcrographs of fracture surfaces from the sulfur - 65 wt. % JSC-1 samples noted in figure 32. Figure 37 is taken from sample 4, a noncycled sample tested at 21 °C (70 °F). Figure 38 is from noncycled sample 6, which is like sample 4 in figure 37, but tested at –101 °C (–150 °F). Fgure 39 s from sample 18, a cycled sample tested at 21 °C (70 °F). Figure 40 is from sample 20, which is like sample 18 in figure 39, but tested at –101 °C (–150 °F). Note that extensve crackng and de-bondng are also seen n the cycled 65 wt. % JSC-1 samples (figs. 39 and 40) when compared to those noncycled samples (figs. 37 and 38)—perhaps, though, not as simply observed as with the samples (figs. 35 and 36) containing the spherical silica particles. Fgure 37. SEM mcrograph of a fracture surface from noncycled sample 4, a sulfur wth 65 wt. % JSC-1 sample tested at 21 °C (70 °F). As mentoned above, no concluson on fracture strength could be determned as a functon of composton or test temperature, yet the cycled samples faled at a load some 5 tmes less than the noncycled. It appears obvous that ths dfference s due to de-bondng of the aggregate partcles wth the 28

Original page 41

Fgure 38. SEM mcrograph of a fracture surface from noncycled sample 6 wth the same composition as in figure 37, but tested at –101 °C (–150 °F). Fgure 39. SEM mcrograph of a fracture surface from cycled sample 18, a sulfur th 65 wt. % JSC-1 sample tested at 21 °C (70 °F). w 29

Original page 42

Fgure 40. SEM mcrograph of a fracture surface from cycled sample 20 wth the same composition as figure 39, but tested at –101 °C (–150 °F). sulfur, essentally leavng a sample full of cracks pror to compresson testng. Samples at –101 °C (–150 °F) behaved lke those tested at 21 °C (70 °F), but those tested at –191 °C (–312 °F), and cycled, had nferor propertes. One must now queston f de-bondng s a consequence of temperature or cyclng or both. Sulfur s undoubtedly one of the most complex elements. It has numerous allotropc forms n the sold, lqud, and gaseous states, as noted earler. Several slghtly dfferent pressure versus temperature phase dagrams have been publshed, and even ts meltng pont stll appears questonable. Consequently, the materal propertes of sulfur, partcularly those at low temperatures, are less than well known. One can, however, make some assumptons n an attempt to shed lght on the observatons made above. Chempruf Concrete1 reports the coefficient of thermal expansion for their sulfur-based product to be ≈1.2 × 10–5/K (6.67 × 10–6/K). The coefficient of thermal expansion for SiO (slca), from 2 s known to be 5.5 × 10–7cm/cm K (3.055 × 10–7/°R).44Sulfur 527.7 °R (20 °C) to 1067.7 °R (320 °C),  and, essentally, slcates, compose the concrete; there s no chemcally reacted bond between them; and there is a difference of two orders of magnitude in their coefficient of thermal expansion. The strain put on a materal as a functon of temperature can be evaluated as follows: ∆L ε = = α∆T . (2) L Here ε s the stran, ∆L s the change n the orgnal length (L), α is the coefficient of thermal expanson, and ∆T s the temperature dfference that mposes the stran. For sulfur, cyclng between 21 °C (70 °F) and –191 °C (–312 °F) gves ∆T = 216 K (381.6 °R), and the calculated stran would be 30

Original page 43

2.6 × 10–2. The calculated stran n SO s 1.16 × 10–4, whch s essentally neglgble but mportant n 2 the sense that two orders of magntude exst between the materals. For metals, the generally accepted transton from elastc behavor to plastc behavor (permanent deformaton) occurs at a stran of 0.002. Assumng ths also apples to sulfur, ∆T for the elastic-plastic transition would be ≈167 K (300.6 °R) and from RT (21 °C, 69.8 °F) corresponds to a temperature of approxmately –146 °C (–230.8 °F). Ths temperature appears reasonable, as an earler study25cycled samples 50 tmes between RT and –27 °C (–16.6 °F), wth compresson test results beng smlar to noncycled, RT samples. Though not cycled, samples compressed at –101 °C (–150 °F) showed no dscernble dfference n propertes or fracture mcrostructure from the strctly RT samples. The data ndcate that a transton occurred whle coolng to –191 °C (–312 °F), and a temperature of –146 °C (–230.8 °F) was suggested. Ths value can only be consdered an estmate, partcularly n vew of uncertan property values, especally at such low temperatures. Also lkely s a functonal relatonshp between temperature and the number of cycles needed to ntate and complete, or nearly complete, de-bondng; heatng and coolng rates may also be a factor. Fnally, t was assumed that the aggregate was pure SO2, when n realty SO2 composed only ≈20%. The rest was JSC-1, a material composed of a number of mnerals, albet mostly slcates. These varables, ncludng volume fracton of aggregate (as well as their size and shape), are all additionally influenced by the inherent sample nhomogenetes whch further compromse accurately determnng concrete vablty. What does appear certan s that the contractng sulfur has a comparatvely poor bond wth the aggregate materal and separates at that nterface rather than fracturng wthn tself. To gan some nsght on cyclng between room and LN2temperatures, samples of sulfur wth 65 wt. % pure SO2 were made (a piece is seen in fig. 41), and placed in quartz test tubes. As expected, no ntegrty was lost when a sample was cycled 20 tmes between RT and –15 °C (5 °F). However, debrs was seen to accumulate after the seventh cycle for the sample tube mmersed n LN2 (–196 °C, or –320.8 °F). By the 20th cycle, the once sold pece had crumbed to free grans of SO2and small pieces of sulfur, as seen in figure 42. Here the only ‘outside’ forces experienced by the sample were gravty and slght movements due to manpulatng the test tube. One also can assume that concrete integrity could be significantly compromised after the first cycle. In retrospect, cycling the samples 80 tmes was certanly excessve. 3.4 Summary of Extreme Temperature Concerns Work was undertaken to evaluate the structural ntegrty of sulfur concrete that was subjected to cyclng between temperatures that mght be expected on the lunar surface. Prevous work showed on strength of samples cycled between RT and –27 °C (–16.6 °F) was not statstcally that the compress dfferent from noncycled samples. In contrast, samples cycled between room temperature and –191 °C (–312 °F) showed at least 5 tmes less strength than those noncycled. Mcroscopc nvestgaton of the fracture surfaces showed clear de-bondng of SO2partcles from the sulfur. The observed de-bondng is attributed to the large differences between the coefficients of expansion of sulfur and aggregate, and it initiates at some yet unknown temperature(s) where the induced strain is sufficient for the sulfur to go from elastic to plastic behavior. A simple test suggested that significant structural degradation initiates after only a few cycles between room and LN2temperatures. Whle de-bondng and poor mechancal behavor are certan, a complete analyss s hampered by lackng materal propertes at low temperatures, 31

Original page 44

Fgure 41. Photograph of a sulfur sample wth 65 wt. % pure SO2. Figure 42. Photograph of a sample like that seen in figure 7 which was cycled 20 times between room temperature and –196 °C (–321 °F). Crumblng of the sample wth free grans of slca s seen. usng partally characterzed aggregate, and usng samples that have nherent defects such as poor aggregate dstrbuton and varable porosty. Such problems wll lkely be exacerbated f sulfur concrete s produced on the lunar surface. 32

Original page 45

1. OVERALL SUMMARY AND CONCLUSIONS Meltng sulfur and mxng t wth an aggregate to form ‘concrete’ s commercally well establshed and produces a materal that s partcularly well-suted for use n corrosve envronments. Dscovery of the mneral trolte (FeS) on the Moon poses the queston of extractng the sulfur for use as a lunar constructon materal, an attractve alternatve to conventonal concrete as sulfur concrete does not requre water. However, the vablty of sulfur concrete n a lunar envronment, whch s characterzed by the lack of an atmosphere and the presence of extreme temperatures, s not well understood. The ntent of the work presented here was to conduct a seres of ground-based experments to gan nsght regard- ng any detrmental effects that the extreme envronmental condtons of the lunar surface mght have on sulfur concrete. Here it is assumed that the lunar ore can be mined and refined and the raw sulfur melded wth approprate lunar regolth to form, for example, brcks. The first issue addressed was the ‘hard’ vacuum environment of the lunar surface and how it mght sublmate away exposed sulfur and degrade any concrete structure. In ths study, small, pure sulfur and two sulfur ‘concrete’ mxtures were prepared and placed n a vacuum envronment (capable of 5 × 10–7 torr) at ≈20 °C (≈68 °F) for 60 days. Perodc weghng of the samples revealed a contnuous weght loss due to the sublmaton of sulfur. Reasonable agreement wth the Hertz-Knudsen equaton was seen over ≈10 days. Subsequent deviation was attributed to nonuniform surfaces, cavity formation, and ncreased exposure of aggregate materal. The sublmaton rate vared from rapd at the hghest lunar temperatures expected to essentally nonexstent at the lowest. Second, blocks of sulfur concrete were cycled between LN2 temperature (≈ –191 °C, or –312 °F) and RT (18 to 20 °C, or 64.4 to 68 °F) to smulate exposure to the extreme cold of the lunar envronment. These, and a smlar set of blocks not cycled, were subsequently subjected to compresson testng at two temperatures, ≈21 °C and ≈ –101 °C (≈70 °F and ≈ –150 °F). No effect of the dfferent compostons or test temperatures could be ascertaned from ether set. However, the compresson strength of the noncycled samples averaged roughly 35 MPa (≈5,076 psi), whereas the cycled samples fractured at about 7 MPa (≈1,015 psi), or approximately one-fifth the load of noncycled samples. The disparity in strength was attributed to significant differences in thermal coefficients of expansion, which promoted crackng. In short, ‘warm’ lunar temperatures mantan mechancal propertes but ncrease sublmaton knetcs, whereas ‘cold’ temperatures mnmze sublmaton effects but degrade mechancal propertes. In concluson, as on Earth, use of sulfur concrete n the lunar envronment as a constructon materal wll requre specal crcumstances. Fnally, t has been suggested that a ‘protectve’ coatng be put on the sulfur concrete. Perhaps, but first, issues of compatibility, safety, resources, and likely other concerns would have to be addressed. 33

Original page 46

REFERENCES 1. GRC, Inc.-Chempruf, P.O. Box 644, Clarksvlle, TN 37040, http://www.chemproofconcrete.com Accessed 3 October 2007. 2. Sullvan, D.: “Acd-Proof Coatng Composton,” U.S. Patent No. 1,808,081, June 2, 1931. 3. Leutner, B.; and Dehl, L.: “Manufacture of Sulfur Concrete,” U.S. Patent No. 4,025,352, May 24, 1977. 4. Vroom, A.H.: “Sulphur Cements, Process for Makng Same and Sulphur Concretes Made Therefrom,” U.S. Patent No. 4,058,500, November 15, 1977. 5. ACI Commttee 548: “Gude for Mxng and Placng Sulfur Concrete n Constructon,” ACI Materials Journal, Farmngton Hlls, MI, p. 314, July–August, 1998. 6. Crck, S.M.; and Whtmore, D.W.: “Usng Sulfur Concrete on a Commercal Scale,” Concrete International, Vol. 20(2), p. 83, February 1998. 7. Czarnecj, B.; and Gllott, J.E.: “Effect of Dfferent Admxtures on the Durablty of Sulfur Concrete Made wth Dfferent Aggregates,” Engineering Geology, Vol. 28(1–2), pp. 105-118, 1990. 8. Head, W. J.: “Fly Ash Sulfur Concrete,” Transportation Engineering Journal, Vol. 107(3), pp. 345– 363, May/June 1981. 9. Khalooand, A.R.; and Ghafouri, H.R.: “Parameters Influencing the Behavior of Sulfur Concrete,” Proc., Intern. Concrete Conf. 92, Tehran, pp. 270–288, November 1992. 10. Lin, S-L; Lai, J.S.; and Chain, E.S.K.: “Modifications of Sulfur Polymer Cement (SPC) Stabilization and Solidification (S/S) Process,” Waste Management, Vol. 15(5/6), pp. 441–447, 1995. 11. Loov, R.E.; Vroom, A.H.; and Ward M.A.: “Sulfur Concrete—A New Constructon Materal,” PCI Journal, pp. 86–95, January–February 1974. 12. Hammons, M.I.; Smith, D.M.; Wilson, D.E.; and Reece, C.S.: “Investigation of Modified Sulfur Concrete as a Structural Materal,” U.S. Army Corps of Engneers, Techncal Report CPAR-SL-93-1, July 1993. 13. Malhotra, V.M.: “Sulfur-Infiltrated Concrete,” Concrete Construction, March 1975. 14. Okumura, H.A.: “Early Sulfur Concrete Installatons,” Concrete International 20, No. 1, p. 72, January 1998. 34

Original page 47

1. Nevn, P.J.: “Assessng Sulfur Concrete Applcatons,” Concrete International 20, No. 2, p. 87, February 1998. 16. Vroom, A.H.: “Sulfur Concrete Goes Global,” Concrete International 20, No. 1, p. 68, January 1998. 17. Vroom, A.H.: “Sulfur Concrete for Precast Products,” Concrete International 20, No. 2, p. 90, February 1998. 18. Taylor, G.J.; Warren, P.; Ryder, G.; Delano, J.; Peters, C.; and Lofgren, G.: “Lunar Rocks,” Lunar Sourcebook, G.H. Heken, D. Vanman, and B.M. French (eds.), Cambrdge Unversty Press, pp. 183–284, 1991. 19. Haskn, L.; and Warren, P.: “Lunar Chemstry,” Lunar Sourcebook, G.H. Heken, D. Vanman, and B.M. French (eds.), Cambrdge Unversty Press, pp. 357–474, 1991. 20. Vanman, D.; Pettt, D.; and Heken, G.: “Uses of Lunar Sulfur,” Second Conference on Lunar Bases and Space Activities of the 21st Century, Wendell Mendell (ed.), Lunar and Planetary Insttute, pp. 429–435, 1992. 21. Casanova, I.: “Feasblty and Applcatons of Sulfur Concrete for Lunar Base Development: A Prelmnary Study,” 28th Annual Lunar and Planetary Science Conference, March 17–21, Houston, TX, p. 209, 1997. 22. Graca, V.; and Casanova, I.: “Sulfur Concrete: A Vable Alternatve for Lunar Constructon,” Proceedings of the Sixth International Conference and Exposition on Engineering, Construction and Operations in Space, Albuquerque, New Mexico, pp. 585–591, Aprl 26–30, 1998. 23. Leonard, R.S.; and Johnson, S.W.: “Sulfur-Based Constructon Materals for Lunar Constructon,” Engineering, Construction, and Operations in Space, Proc. of Space 88, S.W. Johnson and J.P. Wetzel, ASCE (eds.), pp. 1295–1307, 1988. 24. Roqueta, J.; and Casanova, I.: “Manufacture and Propertes of Sulfur Mortar for Lunar Applcatons,” Proceedings of the 7th International Conference and Exposition on Engineering, Construction, Operations and Business in Space, S.W. Johnson, K.M. Chua, R. Galloway, and P. Rchter (eds.), Albuquerque, NM, pp. 851–855, 2000. 25. Toutanj, H.; Glenn-Loper, B.; and Schrayshuen, B.: “Strength and Durablty Performance of Waterless Lunar Concrete,” AIAA proceedngs paper No. 2005-1436. 26. Grugel, R.N.; and Toutanj, H.: “Vablty of Sulfur ‘Concrete’ on the Moon: Envronmental Consderatons,” AIAA proceedngs paper No. 2006-520, 2006. 27. Grugel, R.N.; and Toutanj, H.: “Sulfur “Concrete” for Lunar Applcatons— Sublmaton Concerns,” Advances in Space Research, Vol. 41, pp. 103–112, 2008. 35

Original page 48

1. Jones, L.; and Atkns, P: Chemistry: Molecules, Matter and Change, 4th ed., W.H. Freeman, New York, 2000. 29. Hong, R.E.; Kramer, D.A.: “Vapor Pressure Data for the Sold and Lqud Elements,” RCA Revew, Vol. 30, pp. 285–305, 1969. 30. McKay, D.S.; Carter, J.L.; Boles, W.W.; Allen, C.C.; and Allton, J.H.: “JSC-1: A New Lunar Sol Smulant,” Engineering, Construction, and Operations in Space IV, Amercan Socety of Cvl Engneers, pp. 857–866, 1994. 31. Hertz, H.: “Uber de Verdunstung der Flussgketen, nsbesondere des Queckslbers, m luftleeren Raume,” Ann. Physik, Vol. 17, p. 177, 1882. 32. Knudsen, M.: “De Molekularstromung der Gase durch Offnungen und de Effuson,” Ann. Physik, 4th seres, Vol. 28, p. 999, 1909. 33. Langmur, I.: “Chemcal Reactons at Very Low Pressures. II. The Chemcal Clean-up of Ntrogen n a Tungsten Lamp,” J. Am. Chem. Soc., Vol. 35, p. 931, 1913. 34. McEachern, D.M.; and Sandoval, O.: “A Molecular Flow Evaporatng Apparatus for Measurng Vapour Pressures and Heats of Sublmaton of Organc Compounds,” J. of Physics E: Scientific Instruments , Vol. 6, pp. 155–161, 1973. 35. Langmur, I.: “The Vapor Pressure of Metallc Tungsten,” Physical Review, Vol. II(5) pp. 329–342, 1913. 36. Marshall, A.L.; Dornte, R.W.; Norton, F.J.: “The Vapor Pressure of Copper and Iron,” J. Amer. Chem. Soc., Vol. 59(7), pp. 1161–1166, 1937. 37. Johnston, H.L.; and Marshall, A.L:“Vapor Pressures of Nckel and of Nckel Oxde,” J. Amer. Chem. Soc., Vol. 62, pp. 1382–1390, 1940. 38. Holden, R.B.; Speser, R.; and Johnston, H.L.: “The Vapor Pressures of Inorganc Substances. I. Beryllum,” J. Amer. Chem. Soc., Vol. 70, pp. 3897–3899, 1948. 39. Rosenblatt, G.M.; and Lee, P-H.: “Vaporzaton Knetcs and Thermodynamcs of Antmony and the Vaporization Coefficient of Antimony Single Crystals,” J. Chem. Phys., Vol. 52, pp. 1454– 1464, 1970. 40. Howlett, D.L.; Lester, J.E.; and Somorja, G.A.: “Vacuum Vaporzaton Studes of Lthum Fluorde Sngle Crystals,” J. Phys. Chem., Vol. 75, pp. 4049–4053, 1971. 41. Wang, L.L.; Wallace, T.C., Sr.; Hampel, F.G.; and Steele, J.H.: “Vacuum Evaporaton of KCl-NaCl Salts: Part II. Vaporzaton-Rate Model and Expermental Results,” Met. and Mat. Trans. B 27B, p. 657, 1996. 36

Original page 49

1. Rosenblatt, G.M.; and Lee, P-H.: “Rate of Vaporzaton of Arsenc Sngle Crystals and the Vaporzation Coefficient of Arsenic,” J. Chem. Phys., Vol. 49, pp. 2995–3006, 1969. 43. Inaba, H.; Tachbana, S.; Nagahara, H.; and Ozawa, K.: “Condensaton Knetcs of Forsterte,” Proceedings, Lunar and Planetary Science XXXII, paper 1837, 2001. 44. Weast, R.C. (ed.): Handbook of Chemistry and Physics, 57th edton, CRC Press, 1976–1977. 37

Original page 50

Form Approved REPORT DOCUMENTATION PAGE OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operation and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503 1. AGENCY USE ONLY (Leave Blank) 2. REPORT DATE February 2008 4. TITLE AND SUBTITLE 3. REPORT TYPE AND DATES COVERED Techncal Memorandum 5. FUNDING NUMBERS Sulfur ‘Concrete’ for Lunar Applcatons—Envronmental Consderatons 6. AUTHORS R.N. Grugel 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) George C. Marshall Space Flght Center Marshall Space Flght Center, AL 35812 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) Natonal Aeronautcs and Space Admnstraton Washngton, DC 20546–0001 11. SUPPLEMENTARY NOTES 8. PERFORMING ORGANIZATION REPORT NUMBER M–1223 10. SPONSORING/MONITORING AGENCY REPORT NUMBER NASA/TM—2008–215250 Prepared by the Materals and Processes Laboratory, Engneerng Drectorate 12a. DISTRIBUTION/AVAILABILITY STATEMENT Unclassified-Unlimited Subject Category 29 Avalablty: NASA CASI 301–621–0390 13. ABSTRACT (Maximum 200 words) 12b. DISTRIBUTION CODE Commercal use of sulfur ‘concrete’ on Earth s well establshed, partcularly n corrosve, e.g., acd and salt, envronments. Havng found trolte (FeS) on the Moon rases the queston of usng extracted sulfur as a lunar constructon materal, an attractve alternatve to conventonal concrete as t does not requre water. For the purpose of this Technical Memorandum, it is assumed that lunar ore is mined, refined, and the raw sulfur processed wth approprate lunar regolth to form, for example, brcks. Wth ths stpulaton, t s then noted that the vablty of sulfur concrete n a lunar envronment, whch s characterzed by lack of an atmosphere and extreme temperatures, s not well understood. The work presented here evaluates two sets of small sulfur concrete samples that have been prepared using JSC-1 lunar simulant as an aggregate addton. One set was subjected to extended perods n hgh vacuum to evaluate sublmaton ssues, and the other was cycled between room and lqud ntrogen temperatures to nvestgate ther subsequent mechancal ntegrty. Results are presented from both nvestgatons, dscussed, and put nto the context of the lunar envronment. 14. SUBJECT TERMS 15. NUMBER OF PAGES tu resource utlzaton, sulfur, sulfur 48 lunar envronment, lunar processng, n s tes, lunar sol smulant 16. PRICE CODE concrete, sublmaton, compresson testng, compos 17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION OF REPORT OF THIS PAGE Unclassified Unclassified NSN 7540-01-280-5500 38 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT OF ABSTRACT Unclassified Unlmted Standard Form 298 (Rev. 2-89) Prescribed by ANSI Std. 239-18 298-102

Original page 51

Original page 52

National Aeronautics and Space Administration IS20 George C. Marshall Space Flight Center Marshall Space Flight Center, Alabama 35812