‘0... 4 3‘ funny. _ . u... "or T. .. . 3' .3! ‘J.&§fl‘§ .gUUHfi 515 l‘ ABSTRACT CONCRETE MIXING STUDIES BY GAMMA-RAY ABSORPTION by Carlos A. Zapata The (pality and economy of fresh concrete depend on its ingredients. their proportions, and the homogeneity of the finished product. Selection of the proper ingredients and their proportions are important factors in achieving the desired mix consistency. However, a basic problem is the need for simple and rapid means of measuring uniformity of concrete and of finding the optimum mixing time for the blending operation. A nondestructive, simple. rapid technique was deveIOped for checking the times involved in various stages of the mixing process, together with an indication of the homogeneity of the concrete mixture throughout the batch. In addition, when fresh concrete was designed with strength as the basis for acceptance, the method developed was capable of determining the Optimum mixing time needed to produce maximum strength. The method involved use of a gamma-ray absorption apparatus (densitometer) in which an external radiation source was placed on one side of a drum-type concrete mixer with the transmission recorded on the opposite side. The experimental data were recorded as a function of change in bulk density of the mix. From these data and with the aid of the attenuation law for gamma rays, linear absorption coefficients for concrete materials were obtained. These results were of considerable assistance in evaluating and predicting overall performance of the gamma-ray densitometer. Carlos A. Zapata This technique's ability to detect important variations involved in the concrete mixing operation was confirmed by a concrete-mixing study using sodium chloride. Dry sodium chloride was mixed with the concrete materials. Several samples were taken at successive time intervals during the entire cycle and titrated with silver nitrate. The general distribution of the results obtained followed a regularity very similar to that found in the gamma-ray data, especially during the clumping period. Furthermore, the end of this clumping period also appeared to be the end of the variability of sodium chloride values before falling within the range of experimental error. In both methods this period ranged between 60 and 85 seconds. CONCRETE MIXING STUDIES BY GAMMA RAY ABSORPTION By Carlos A. Zapata A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemical Engineering 1961 To My Wife , Sarita and My Mother, Tulia ii ACKNOWLEDGEMENT The author particularly wishes to thank J. W. Donnell, professor of chemical engineering, R. W. Ludt, professor of chemical engineering, and H. Rubin, professor of statistics, of Michigan State University, for their constructive criticism and valuable suggestions. The presentation of this paper has been made possible through the cooperative efforts of many persons of the Research Laboratory Division of the Michigan State Highway Department. In particular, thanks are due to E. A. Finney, Director, and to B. W. Pocock, supervisor of the Isotopes Research section, for their contributions, and to others for graphic presentation and editorial review of the manuscript. iii CONTENTS Introduction ___________________________________________ Radiation History ....................................... Interaction of Gamma-Rays with Matter....---_------_..---_--___-------.. Fundamental Law of Radioactive Disintegration -..--_--- Operational Theory for Gamma Rays- .............. The Gamma-Absorption Measuring System-..” ...... The Radiation Source -- ----”——-------—_—- The Solid Angle Subtended by the Counter ........... Background Radiation Effects of Scattered Radiation ...................... Equipment and Materials --- Experimental Procedures and Results Density Calibration Curve The Gamma Ray Trace--..--.___-__----__-.._- Laboratory Tests _____ Field Tests ........ .. Photographic Study of the Mixing Process Mixing Time Study with Sodium Chloride -_ ASTM Methods _ -_--__-_-------_---- Discussion of Results ................ Summary .............. Conclusions _______________________________ Suggestions for Further Work _... ....................... Appendix .................... Computations -_- ............. Radioactive Disintegration .......... Bibliography .__._ iv 66 67 69 71 INT RODUC TION Although a great deal of work has been done in sampling, mixing time, and testing of freshly-mixed concrete, most efforts have been concentrated in evaluating concrete uniformity in terms of such prOperties as compressive strength, workability, and durability by applying well-known standard pro- cedures. A strong, durable concrete is obtained by correctly proportioning and uniformly blending the required quantities of water, cement, and aggregates (sand and gravel) throughout the batch. The relation of water to cement, usually referred to as the water-cement ratio, affects the compressive strength and durability of the concrete mixture. The higher strengths frequently are obtained with the higher water-cement ratios, but too much mixing water reduces concrete strength and durability. Measurement of the effects of water content on con- crete is possible by the standard slump test (45). Since this test gives a good indication of the water content of the batch, it provides a basis for control of consistency or stiffness of the fresh concrete mixture. Usually, the limits for this test should be maintained within 1 to 3 in. of slump. When air—entrained cement is used, care should be taken to keep the air content of the freshly mixed concrete within a given limit (usually 3 to 6 percent by volume). Should the air content be higher than 6 percent, concrete strength will be reduced'without any gain in durability. If the air content of the concrete is less than 3 percent, the mixture may offer poor resistance to alternate freezing and thawing after the concrete has hardened. Measurement of the air -2- content of freshly mixed concrete is usually included in any program for testing uniformity of concrete (5, 14). Workability or plasticity of the concrete is considerably influenced by the quality of the mortar paste (cement, sand, and water), which fills all the voids and keeps the coarse aggregate sufficiently separated so that it may move freely during the mixing period. Consequently, water and cement content of mortar and the percentage of coarse aggregate in the concrete are usually determined in any mixer performance test (5, 14). Finally, concrete quality is also influenced by the amount of mixing required to produce adequate homogeneity. With given concrete materials and similar conditions of mixing, 'a longer mixing time will reduce the air content and the strength of the finished product. Some specifications require that minimum mizdng time for a standard drum-type mixer should be no less than 1 minute for mixers having a capacity of 1 cu yd, with an increase of 15 seconds in mixing time for each additional 1 cu yd of concrete mixed when larger mixers are used (5). But when freshly-mixed concrete is used with strength as the basis for uniformity, careful attention must be given to mixing time between batches. Despite the fact that all these conventional tests are time—consuming and destructive in nature, very little effort has been concentrated toward developing rapid and nondestructive methods to evaluate uniformity of concrete and Optimum mixing time. For instance, it is well known that elements present in ordinary concrete show different radiation absorption prOperties when they are exposed to gamma rays (32, 33), yet little attention‘has been given to the use of radio- activity in testing the uniformity of fresh concrete. -3- A recent investigation using radioisotopes for study of mixing efficiencies in ready-mixed concrete was conducted by Overman and Rohrman (31). This study examined mixing times for heavier and less homogeneous materials, using a small cement mixer and colloidal gold 198 as the tracer. This work has not yet been fully reported. Another investigation is being conducted by the Georgia Institute of Technology for the Atomic Energy Commission (41). The Georgia study is concerned specifically with determination of cement content by neutron activation. Samples of cement mortar of known cement content are activated by the "Van de Graaff" machine and the calcium 49 produced is counted in the Penco 100 channel analyzer. Count rates with energies above 2. 0 million electron volts (mev) have been recorded for each known cement content. Some inconsistencies have been observed in these results. R. M. Main (13) has recently pointed out the limitations of radioactive tracer techniques for rapid determination of uniformity of mixing. Random sampling variation, sample size, and tracer technique applied to quality control are discussed in his paper. The tracer technique involves adding a small amount of radioactive compound to the material being mixed, and after the blending Operation has proceeded, taking several samples to determine their content of tracer. Other investigations have been concerned primarily with the radiation- shielding properties of hardened concrete (32, 33). Gamma-ray absorption techniques have been used successfully for measuring density of Opaque materials and solids concentration in fluids (7); for measuring density of liquids flowing inside a pipe (21); and for controlling thicknesses of solid materials and liquid levels inside tanks (17, 35). -4- The measurement of density is based on the known interaction of gamma rays and the orbital electrons of atoms comprising the absorbing material. Another application of the gamma-ray technique, based on back-scattering radiation, is the determination of density of soils and similar granular systems (11, 10, 36). The present investigation was intended to evaluate the ability of an external gamma—ray system to detect significant variations during concrete mixing, and to determine, if possible, the optimum mixing time. The program was divided into three major parts. 1. The first was acquisition of a basic understanding of the nature of the problem and any relevant theory associated with it. 2. The second phase was verification in the laboratory of those results found in the literature which bear on the problem, such as attenuation character- istics and absorption coefficients of concrete materials. For this, a given com- position of concrete materials was mixed in a standard drum-type mixer of 137 was used, 1. 5 cu ft capacity. As an external radiation source 5 me of Cs together with a standard scintillation counter and a recorder. In addition, to verify the results found in terms of Optimum mixing time by using the gamma- ray absorption method, other techniques were used, such as photography, tracing sodium chloride in the mixture, measuring compressive strength, determining air content, and conducting a slump test. 3. Finally, the design problems were considered for a permanent installation of the gamma-ray system to test uniformity of large batches of concrete (3 to 8 cu yd), and some eXperimental results were obtained in the field. RADIATION HISTORY The process of giving off particles or rays from the nucleus of unstable atoms is known as radioactivity, sometimes termed nuclear radiation, dis- integration, or radioactive decay. In 1896, Henri Becquerel found that compounds of uranium were capable of spontaneously emitting radiations that were able to affect a photographic plate in the same manner as if the plate had been exposed to light. He also found that such radiations could ionize air. In 1898, Pierre and Marie Curie carried on Becquerel's work and discovered the element radium, which was very much more active than uranium itself. Shortly afterward a new compound, polonium, was produced which was also active, and since that time many radio- active elements have been found. Four series Of radioactive elements have been identified deriving from uranium, thorium, actinium, and neptunium. The first three decompose through a number of successive radioactive changes, emitting radioactive particles until a non radioactive end product is finally obtained. Natural lead is a mixture of the three end products plus a fourth isotOpe, Pb204. The discovery of the neptunium was a result of many experiments with the atomic pile. This series has an isotope of bismuth as its end product. The radiations emitted from radioactive substances are of three types, designated as alpha particles (c( ), beta particles (,6), and gamma rays (7‘ ). More information related to radioactive hazards from these radiations are presented in the appendix. INTERACTION OF GAMMA-RAYS WITH MATTER Because continuous detection and measurement of gamma rays is a major subject in the present experimental work, some understanding of the mechanism on which these measurements are based is desirable. Practically all instruments available for detecting and measuring natural or artificial radiation depend on the ionization produced directly or indirectly through the interaction of radiation with matter. The ionizing effect of charged particles, such as alpha and beta rays, is produced directly in the interaction of such particles with the atoms of the traversed matter. But gamma rays do not produce ionization in matter directly. They interact with matter by pro- ducing secondary ionization which takes place in three ways, designated as photoelectric, pair production, and Compton effects. In the photoelectric effect, an atom irradiated with a gamma photon absorbs the radiation by ejecting an electron (Fig. 1). The gamma photon is completely absorbed, resulting in the excitation or ionization of atoms of the matter. This mechanism of absorption of gamma rays is important only for photons with less than 0. 1-mev energy. When the energy of gamma radiation is greater than 1. 02 mev, the mechanism of pair production is possible. The gamma photon disappears and an electron pair is created (Fig. 2). For gamma radiation with energy ranging between 0. 1 and 10-mev, the mechanism of Compton effect becomes important. This effect can be treated as an elastic collision between a gamma-photon and an electron. The photon -5- .893 no «vote ”-39:00 .m enema ZOCFumJN flouuc I. ZOFOIQ k oucwaKOm >u2 0.9V “V >U2 ..O ZOPOIa k .PZNQUZ. .893 no «vote none-60.5 ham .m magma ZOEtmOQ >u§ «0.. AU mmme- x I. 3 ZOchUUJu .EBm no Soto oanoBoeoam .H 0.263 no >u: tsvu ’ zopoxax -8— rebounds in some direction, making an angle ¢ with respect to the original path. The electron recoils at an angle 9 (Fig. 3). The Compton effect will be the only important process in the present experimental work. FUNDAMENTAL LAW OF RADIOACTIVE DISINTEGRATION In performing gamma-ray experiments such as those associated with the present project, it is important to compute the amount of radioactive source and activity involved. This can be done by applying the fundamental equation of radioactive disintegration: _ 0.69t N =Noe t9: (1) where N = decayed intensity in millicuries (me) N0 = original intensity in me t = elapsed time since irradiation t ,2 = half-life of the particular radioactive substance. One of the characteristics of a radioactive decay process is half-life, defined as the time required for radiation to decrease to half the original value. The half—life of each isotOpe is constant and cannot be changed by currently known electrical, physical, or chemical forces. In the case of gamma-ray emitters (which were used in the current experi- mental work), strength is often given in terms of curies. The curie is a unit of radioactivity defined as the quantity of any radioactive material giving 0 disintegrations per second. The millicurie (mc) which is one- -3. 7 x 101 thousandth of a curie, and the microcurie (no) which is one-millionth of a curie, are equivalent to amounts of radioactive material giving 3. 7 x 107 and 3- 7 x 104 disintegrations per second, respectively. Practical calculations of the amount of radiation source and its activity are illustrated in the appendix. OPERATIONAL THEORY FOR GAMMA RAYS The gamma—ray-absorption method utilizes the effect resulting from eXposing material to incident radiation. This method is based on the principle that gamma rays are absorbed in prOportion to the thickness or density of the interposed material. When a collimated or narrow beam of gamma rays hits a sample of concrete material, the beam is attenuated according to the funda- mental equation: 1 = :0 e' “(PL (2) where I = intensity of radiation after passing through the absorbing material Io = intensity before passing through the absorber at = the mass absorption or attenuation coefficient ,0 = density of the material L = acmal radiation path length. The second fundamental law is that the radiation intensity, I0 , is inversely proportional to the square of the actual radiation path, L . In the literature alpha is often expressed as: °C= #- (cmZ/g) (3) where it = "linear attenuation or absorption coefficient, " in units of (cm-1). ,0 = density of the material in (g/cm3). Equation 2 is also written as: -ccx =e (4) "attenuation factor or transmittance" £— total mass of absorbing material per unit area in (0mg). where The coefficient alpha depends only on the composition of the absorbing material and not on the bulk density. The linear coefficient, u, depends on the chemical and physical states of the absorbing material. It is proportional to the sum of the Compton, photoelectric and pair production effects. -10- -11- For the elements present in ordinary concrete, radiation absorption is mainly due to the Compton effect for gamma rays with energy ranging from 0. 2 to 3.0 mev. This may be observed in Fig. 4 from data given by Overman and Clark (18). Values of u are shown for the three mechanisms of gamma-ray absorption by aluminum (or by concrete) as a function of gamma-ray energy. Typical curves of the absorption coefficient, a, as a function of gamma-ray energy (in mev) appear in Fig. 5 as reported in the literature (1). 10 ‘ Absorption coefficient, cm" 8 0.01001 0.1 1 I) Gamma may. Mu Figure 4. Values of u for three mechanisms of gamma- ray absorption as afunction of gamma— ray energy, from Overman and Clark (18). meAn Assoapnou cocmcusm, )1. fun") CONCRETE AIR (3:103) o ‘u 2 3 4 5. o 1 o GAMMA-RAY turner (Mtv) Figure 5.‘ Typical curves of absorption coefficient, )1, as a function of gamma-ray energy (1). THE GAMMA-ABSORPTION MEASURING SYSTEM The different functional parts of the gamma absorption apparatus (Fig. 6), arranged for the purpose of detecting important variations during a concrete mixing Operation, include: 1) the gamma ray source, 2) the scintillation counter, 3) the ratemeter, 4) the amplifier, and 5) the recorder. While the mixing Operation is in progress, changes in the physical char- acteristics of the concrete materials are continuously exposed to a narrow beam of gamma rays. These gamma rays are absorbed in proportion to the thickness or density of the interposed material according to Equation 2. The attenuated rays are detected and converted into electrical signals by a scin- tillation counter. These random signals or pulses are either converted into an average count rate in a ratemeter, or amplified to produce a continuous record of those variations involved in the mixing process. For the purpose of evaluating and predicting the overall performance of this gamma ray densitometer, the following equations of Alcock and Ghosh (17) are of great assistance: ed‘K/e Ax: «(3.0x1063a 81E) (5) and Ax eix" X = x¢c(3.0x1061n81E) (6) where A1: = resulting error in measurement of absorbing material's mass in mg per cm2 T = relative mass error at =mass absorption coefficient in cm2 per mg S = source strength in millicuries (mc) 1 = number of gamma photons produced per disintegration .0. = solid angle of radiation subtended by the detector 1' = time elapsed in detecting a given amount of radiation E = detector efficiency. -14- RATEMETER AMPLIFIER SCINTILLATION Recoaosn GAMMA RAY M'XER COUNTER SOURCE Figure 6. Gamma-ray absorption measuring system. RADIATION SOU RCE SODIUM IODIDE CRYSTAL (SCINTILLATION COUNTER) Assoaesn Figure 7. Detecting scattered radiation. -13- Equations 5 and 6 indicate that the absolute and relative mass errors will have a minimum value when 1 , n , S , r , and E are maximum. Consequently, the overall performance of the present measuring system can be optimized by finding the proper relationship among the fundamental factors involved in the experiments. These factors may be separated into two main types: 1. Those included in the analytical Equation 6, and 2. Those associated with environmental conditions. The major variables involved in Equation 6 are: a) strength of the radiation source, b) actual radiation path of the gamma rays, 0) solid angle of radiation subtended by the scintillation counter, d) counter efficiency, and e) counting time elapsed while radiation is falling on the detector. On the other hand, among those factors associated with environmental conditions, particular attention should be paid to background radiation which comes from cosmic rays or from other sources of radioactivity in the vicinity, and scattered gamma rays which come from surrounding environment or from parts of the apparatus not directly in the absorption path. As a practical example, the design problems of the present gamma ray densitometer may be considered. The Radiation Source According to Equation 6, by increasing the strength of the source, relative mass error is reduced, and consequently improved measurements are obtained. By doing so, more signals, pulses, or counts per a given time are collected in the ratemeter or recorder. In other words, the larger the radiation source, the more counts are detected, and consequently the more accurate is the reading. -17.. Since the random nature of radioactive disintegration obeys the law of statistics, all instrument readings are in error according to the following formula: CI = 1.6 ‘ (7) :n where f = relative reliable error commonly used in radiocounting and some- times called "nine-tenths error" because there are nine chances out of ten that the error will be smaller N = total number of observed counts. For instance, an instrument reading of 10, 000 counts per minute gives a reliable error of 1.65 percent. This statistical error must be added to the other errors introduced by the ratemeter, amplifier, and recorder. The major problems of enlarging the radiation source in order to increase accuracy are greater costs for both the source and the shielding to reduce the amount of external irradiation to permissible levels. As was pointed out already, statistical fluctuations are inherent in radio- counting. So, when the radiation intensity is lowered sufficiently (i. e. , N in Equation 7 becomes small) by the concrete materials under test, the relative reliable error or noise signal from such fluctuations becomes important. If the total number of observed counts becomes too small, the typical electrical signal being sought may be hidden by the statistical noise signal and not observed in the counter or recorder. Therefore, it is desirable to know the Optimum radiation path giving the maJdmum signal-to-noise ratio. This optimum path, L: is found by differentiating Equations 5 and 6 with respect to of and x and equating the results to zero. In each case the relationship at x = 2, yields the maximum signal-to-noise ratio which corresponds to a transmittance of -13- 1 — ——2 = 13. 6 percent. Similarly, the Optimum radiation path, L/, in terms 6 = = 37 percent. _1__ I() of maximum absolute signal for density change occurs when the transmission is .1. I0 6 For the concrete materials used in the present research program, approxi- mate values of L/were computed as listed in Table I. The Solid Angle Subtended by the Counter Referring back to the Equation 6, by enlarging the solid angle ,n. enclosed by the counter, improved readings are obtained since more counts are collected. But this condition is particularly applicable to those systems where the mass of the absorber is sufficiently large that surface effects associated with scattered radiation are completely negligible. In the present study the situation is different. In fact, the geometry of the absorber (mixer plus concrete materials) is so closely connected with multiple scattering that the solid angle 11 must be adjusted to the desired precision in order to keep the large amount of scattered radiation from entering the detector. Background Radiation In practice, if high accuracy is desired, the counting rate should be corrected for background radiation due to the presence of naturally radioactive materials and cosmic rays. This is done by subtracting the background rate from the recorded counting rate. The background rate is determined by taking the count or reading shown in the meter when no radiation source is nearby. Effects of Scattered Radiation As stated before, scattered radiation associated with the geometry of the present system may increase the actual error in radiocounting if the solid angle 11 is not properly adjusted (Equation 6). TABLE I I OPTIMUM PENETRATION OR RADIATION PATH, L, X* FOR MINIMUM ABSOLUTE MASS ERROR, Ax, AND RELATIVE MASS ERROR, A x Approximate Penetration, L’ (in.) Material ForI— =1 =13.6% Forl =l=37.o% e2 e ~ Water 10 5 Cement 8 4 Sand 7 3 Gravel 7 3 Dry mix 6 3 Fresh concrete 4 2 * Results obtained by using linear absorption coefficients of concrete mix materials as listed in Table IV. Radiation source: 5 me C3137. -20- Since bulk density of concrete materials changes over a wide range of values while the mixing Operation is in progress, the attenuation coefficient u should vary quite markedly (Equation 3). This result is to be expected because the mass absorption coefficient remains nearly constant under the present experimental conditions. If quantitative measurements of density variations in the mixture are required, care must be exercised to avoid the measurement of scattered radiation which would tend to decrease the apparent value of the coefficient u. Three types of scattered radiation are associated with the geometry of the measuring system (8): a) Scattered radiation from the surroundings, b) Gamma rays that hit parts of the system not directly in the sbsorption path, and c) Gamma rays that are scattered through a small solid angle (less than 0. 5 steradian) but still measured by the detector. The three types of scattered radiation are illustrated in Fig. 7. The scattered radiation from the surroundings or path (A) and the radiation outside the cone of gamma rays SD'D" or path (C) can be reduced to a negligible amount by collimating of both the source and the scintillation counter. Since all gamma rays scattered through an angle inside the cone SD'D" as shown in Fig. 7 (i. e. , path B) cannot be prevented from reaching the counter, their effect on the attenuation factor of Equation 4 may be illustrated by a simple model as shown in Fig. 8. -21- 7° PHOTON AIR VOID M ABSORBER »‘:.: :11: '2'}:- =..'.':.°:§o°_ . .. 7 PHOTON I. Figure 8. Gamma-ray scatter before (tap) and after (bottom) mixing Operations. . It‘ll. ll '1 {Into .1 l I .lllllmkqlallII' .III II I'llllllll II ll‘lllll | -22- Consider an air void or a small space between aggregate particles char- acteristic at the beginning of the concrete mixing (Fig. 8). Like many others, this air void must be almost completely filled by cement paste, fine sand, gravel, water, or any combination of these, while the material is being mixed. This physical condition, if attained, is one of the important qualities desired in a prOperly designed concrete mix. When no absorber is interposed in the radiation path L (Fig. 8), then from Equation 2, «(,0 L = 0, since the absorption coefficient for air is negligible; I therefore ‘— = . When the absorber is passing through the actual radiation Io beam, the attenuation factor varies continuously. Thus, from Fig. 8: . h 8111 6 = 3%; I L (8) 01' x: y__L_t':_ h (9) By substituting the preceding equation into Equation 4: at L19 i : e + (10) Io The data collected in Table II and plotted in Fig. 9 are based on n values as listed in Table IV. A special case for L = 20 cm and h :- 5 cm was selected to illustrate the effect of the concrete materials upon the attenuation factor when gamma rays scattered inside the cone SD'D" (Fig. 7) are detected by the scin- tillation counter. TABLE 11 1_ EFFECT OF ABSORBER HEIGHT, y, ON ATTENTUATION FACTOR, 10 FOR GAMMA RAYS SCATTERED THROUGH CONCRETE MATERIALS“ Material ly=1cm1y=2cmly=3cmly=4cmy=5em Water 0.75 0.56 0.42 0. 31 0. 22 Gravel or sand 0.63 0. 39 0. 25 0. 15 0. 10 Cement 0.68 0.46 0.32 0. 22 0.15 Cement paste 0. 71 0. 51 0. 36 0. 26 0. 18 Dry mix 0.60 0.36 0.22 0.13 0.07 Fresh concrete 0.50 0.24 0.12 0.06 0.03 * Based on p values listed in Table IV. 43.338 39:58 25 awaoufi venous-3m wheat-«Sam» no“ Eamon nonuomna CO corona a me .333 aoaaeseofl-e- .m one-ME o / ll /. a I I 11 111/ :1 I I I . /. / 11 ’1 all /. / /.. 2.. // 0/ / a 7/ 5.... L I“ u...n(a .PZUZuU PZUZHU oz(¢O L x.’ >¢O L 5528 :85- L r EQUIPMENT AND MATERIALS The essential features of the gamma-ray densitometer, as described here, are shown in Fig. 10. 1. Scintillation Detector Model D85, Nuclear-Chicago Corp. This device can be used for detecting and measuring the energy of alpha, beta, or gamma radiation. For counting gamma rays only a crystal of sodium iodide (thallium activated) is adapted as a main accessory for such detection. The energy of the incident gamma-ray photons is converted into a prOportional amount of visible-light energy. The final multiplied output pulse is proportional to the energy of the incoming radiation. 2. Oscillograph, Double-Channel, Direct Inking, Model BL-202, Brush Development Co. This instrument is designed for malng instantaneous, permanent chart records of a wide variety of electrical phenomena. It provides a chart drive mechanism for pulling radially ruled paper at constant speed under the point of a pen resulting in a record showing variation of the phenomenon under study with time. 3. Universal Analyzer Model BL-320, Brush Development Co. This instrument is a self-contained a. c. Wheatstone bridge, voltage amplifier, discriminator, and d. c. power amplifier. It is used with the direct- inking oscillograph. 4. Count Ratemeter, Model 1620B, Nuclear-Chicago Corp. ' This ratemeter is designed to convert random pulses received from an external radiation detector into an average count rate for presentation on a panel meter and an optional external recorder. -25- dob-om eon-3:3 C. use .3wa 8 £30306 conga-now 3 63.6 «page 3. .2 93m: we -27- 5. Portable Sealer, Model 2800, Nuclear-Chicago Corp. The sealer is a self-contained electronic sealer and power supply for use with nuclear radiation detectors. It may be used with power furnished from the storage battery in its case, or it may be used as an a. c. -Operated instrument. 6. Chart Take-up Drive, Model BL-933, Brush Development Co. This take-up drive is used with the double channel Brush oscillograph to reel recorded chart paper into a form which is readily stored. 7 . Safety Equipment: a. A radiation monitor surveys the radiation intensity in the working area. For persons who are exposed to small-intensity radiation from external sources, a tolerance dosage of 0.3 It. per week is permissible. b. A film badge provides a permanent record of exposure to radiation by measuring accumulated dosage over one or two week periods. 8. Radiation Sources: a. Five millicuries of Cesium 137 with a half-life of 33 years, results in about 2 percent loss in activity and count rate per year. This loss may be corrected for by use of a calibration curve. The source is enclosed in a hermitically-sealed metal capsule and stored in a portable lead-lined case, which absorbs most of the gamma rays produced. The lead case and capsule are sufficiently small and light to be transported readily. b. Forty-five millicuries of Radium 226 and Beryllium with a half-er of 1620 years so that no correction for source decay is necessary. The sealed source is also stored in a lead-lined case. EXPERIMENTAL PROCEDURES AND RESULTS It has been mentioned that these investigations are concerned with two important factors which are involved in practically every mixing process: 1) to trace the uniformity of concrete by detecting significant variations during the mixing process, and 2) to find, if possible, the optimum mixing time while the blending operation is in progress. In an attempt to meet these objectives, a new technique was selected under the assumption that the combined effect of local changes in bulk density of the mixing materials and mixer design could be traced continuously in a simple gamma ray densitometer as described previously (Fig. 6). Thus, the various stages involved in the experimental work are discussed in the following order: a. Concrete materials, mixing conditions, and a Specific mixer for laboratory experiments. b. Proper selection of radiation source, detector, and collimated arrange- ments related to the geometry of the prOposed measuring system. c. Attenuation properties of concrete materials as a function of their density and absorption coefficients. d. Main features of the resulting gamma-ray trace as a function of mixing time. e. Other techniques to verify those results obtained with the gamma ray densitometer. This stage includes a photographic study of the mixing process, a mixing time study with sodium chloride, and ASTM methods for compressive strength, air content, and slump tests. -28- -29- The fresh concrete used throughout these laboratory experiments had the following composition: lb Air-entrained portland cement 2E Natural sand (dry) 64. 0 Gravel: a) Coarse: 3/4 in. to 1/2 in. 23. 3 b) Medium: 1/2 in. to 3/8 in. 23. 3 0) Fine: 3/8 in. to N-4 23.3 Water 11.33 Total weight 169. Total volume about 1. 15 cu ft Unit weight (density) about 147 pcf Thirty concrete batches with this composition were tested after a mixing time of 1. 5 minutes in a standard drum type mixer of 1. 5 cu ft capacity, in terms of within-batch variations in unit weight, slump, air content and com- pressive strength. The following results were obtained in tests conducted according to ASTM methods: Slump 2 to 3 in. Air content 3 to 6 percent by volume 7-day compressive strength 2500 to 3000 psi Unit weight 145 to 147 pcf A commercial batch mixer was used in this study (Fig. 10). It has three steel paddles placed in the drum to aid in the mixing operation. The mixer is ' designed to run by electric power and to operate at 24 rpm. In planning the mixing Operation a standard procedure was desirable in order to form a basis for comparative observations and also to check the results. The following technique was used: 1. The weighted amount of sand and gravel was charged into the mixer and blended for about 5 seconds. -30- 2. The measured quantity of cement was added and all the dry concrete materials mixed for 5 additional seconds. 3. After introducing necessary water, the revolving continued until a mixing time of 1. 5 minutes had elapsed. In mixing concrete materials at least two important features deserve particular attention: a) The mixer design, which causes local changes in bulk density of the materials before the mixing water is being thoroughly distributed; this is the period of mechanical clumping that is characteristic at the beginning of the mixing cycle. b) The large volume of air voids or spaces between aggregate particles that must be almost completely filled by the cement paste in order to achieve the qualities desired in a prOperly designed concrete mix. These continuous changes in the physical characteristics of the mixture during the blending process were exposed to a collimated beam of gamma rays. Impulses picked up by the scintillation counter were fed into a ratemeter, then to an amplifier, and from there to a recorder (Fig. 6). An ideal gamma source for this study should decay mono-energetically in the range of 0. 3 to 1. 5-mev, have a long half life, and be available at low cost. Cesium 137 and Cobalt 60 are good radiation sources and are available at reasonable prices. Cesium 137 has a radioactive half-life of 33 years and emits monoenergetic radiation of 0. 66 mev. Cobalt 60 decays with a half life of 5. 3 years, and for all practical purposes may be considered to emit mono- energetic radiation with a mean energy of 1. 25 mev. -31- An encapsulated, collimated 5-millicurie Cesium-137 source and a scintillation counter equipped with a crystal of sodium iodide (thallium activated), were used for laboratory investigations. A scintillation counter was chosen because of its versatility and high gamma ray sensitivity. This detector can be quickly adapted also to count alpha rays (zinc sulphide crystal) or beta rays (antracene crystal). Collimation was possible by inserting the source capsules as far as possible into a 1/2—in. hole drilled axially in a lead cylinder to within 2 in. of the bottom, holding the capsules near the bottom of the hole by inserting paper stuffing, corldng the hole, and maintaining the cork in position with friction tape. A cone of gamma rays rather than a parallel beam was thus produced, its intensity being very high along the axis in front of the cork, and falling off rapidly at increasing angles from the axis. Similarly, the solid angle of acceptance of the scintillation counter was collimated from its normal 2 pi steradians (50 percent of the total angle) to as narrow a cone as possible equivalent to approximtely 0. 5 steradians. This was done by wrapping five thicknesses of 1/8-in. lead sheet around the tube, projecting approximately 5 in. beyond the crystal. A lead circular plate with five 1/4-in. holes was then inserted in front of the collimated tube. In this way, the effect of scattered radiation associated with the geometry of the concrete mixer was greatly reduced and the overall performance of the measuring system was satisfactory. The critical dimensions of this gamma absorption apparatus are shown in Fig. 11. These dimensions were established after a careful series of preliminary observations in the laboratory. SIDE VIEW PROJECTED RADIATIm CONCRETE MIXER 34" l FLOOR Figure 11. Critical dimensions of the gamma absorption apparatus. To check the overall performance of this system, the radiation source was placed at various known distances from the counter and a set of averaged readings were taken at each position. The ratemeter and the continuous chart recorder were used simultaneously during the measurements. The resulting calibration curve is shown in Fig. 12. The same readings corrected for back- ground radiation and based on the definition of a milliroentgen as described in the appendix can be used to check instrument readings against calculated intensities . RATEMETER READINGS (COUNTS PER MINUTE) Iqooo gooo qooo /// gooo gooo 0 IO 20 30 4O HEIGHT ABOVE BASE LINE (MM) Figure 12. Laboratory calibration curve for ratemeter readings as a function of height above base line on a trace. DENSITY CALIB RATION CURVE As stated before, quick changes in bulk density of concrete materials ‘ take place at the beginning of the mixing cycle. Consequently, according to Equation 3, the linear attenuation coefficient u should either decrease or increase as the absorbing material becomes less or more dense, respectively. So, after the measuring instruments were adjusted and put in good operating condition, the first step was to establish a density calibration curve for the concrete materials. This was done by placing each weighed ingredient separately in a sheet metal container of known volume, and laying the absorbing material between the source 'and the detector. Intensity of radiation was measured before and after the collimated beam of gamma rays passed through a thickness of 24 cm of absorber. The transmitted radiation was measured in counts per minute by using a standard portable sealer in conjunction with the scintillation counter. The actual readings corrected for background radiation are listed in Table III and plotted as a function of density in Fig. 13. The bulk density was computed from the given amount of material used and the known volume of the container. From these data and with the aid of Equation 2 the linear absorption coefficients, [1, for the concrete materials were computed. The results are presented in Table IV and plotted in Fig. 14. -35- -35- Io3x 32 I Q\WATER as ° 24 \ Ill *5 \ z 20 5 X/CEMENT o: u 16 a. m '2 \ 3 '2 O \I'GRAVEL 3 A“ l a SAND rDRY MIX . \ \ FRESH CONCRETE -/?\ l 1 1 O 0 20 4O 60 60 I00 I20 I40 I60 DENSITY (PCF) Figure 13. Density of concrete material versus counts per minute. LINEAR ABSORPTION COEFFICIENT, )1 (CM") 0 O O -37- density of the concrete material. . IS I FRESH CONCRETE? ... mm“.-. _ I I4 .___ - , / ' f /DRY MIX l2 ~— - > \CRAVEL ,m ___L_-,‘__, I .00 06 ——--T#~#* ' .04 we . 02 I .00 O 30 60 90 l20 DENSITY (PCP) Figure 14. Linear absorption coefficient versus ISO -33- TABLE III GAMMA-RAY ABSORPTION DATA FOR CONCRETE MATERIALS“ [ Reading counts er minute Mean Material ( p ) _ __ I No. 1 No. 2 No. 3 No. 4 X X-lb Water 30, 429 30, 238 30, 121 30,091 30,220 28, 920 33, 125 32,939 33,055 33, 201 33,080 31,780 19,428 19,079 19,046 19,018 19,143 17, 843 Cement 18,902 18,071 18,159 18,173 18,326 17,026 Sand 11,453 11, 301 11,625 11,651 11,508 10, 208 12,057 12,315 12,366 12,251 12,247 10,947 11,431 11,224 11,566 11, 341 11,391 10,091 12,029 12,162 12,196 12,242 12,157 10,857 Gravel 11,361 11,251 11,424 11,467 11,376 10,076 11,915 12,119 12,001 12,221 12,064 10,764 12,104 11,934 12,069 11,975 12,020 10,720 Dry Mix (cement, sand, 9, 262 9, 449 9, 331 9, 385 9, 357 8, 057 and gravel) 4,309 3,993 4,132 4,015 4,112 2,812 Fresh concrete 3, 542 3, 575 3, 501 3, 399 3, 504 2, 204 3,782 3,728 3,812 3,894 3,804 2,504 3,935 3,981 3,963 3,804 3,921 2,621 * Average initial intensity, I0 = 174, 000 counts per minute; average background radiation, 1b = 1300 counts per minute. -39 - TABLE IV LINEAR ABSORPTION COEFFICIENTS AND DENSITIES OF VARIOUS MATERIALS* Linear Absorption Material Coefficient, Density, )0 (pef) u (cm-1) “Huer 0°071 02.4 0.074 Cknnent 0.095 86.7 0.097 88.0 . 0.118 104.0 Sand 0.115 100.0 0.118 110.9 0.115 107.5 Gravel 0.118 111.0 0.110 108.0 0.110 103.0 Dry Mi" 0.128 117.9 (cement, sand, gravel) 0.172 112.0 Fresh concrete 0° 182 144- 0 0.170 144.0 0. 175 143. 5 * Densities obtained by weighing 0. 58 cu ft of absorbing material. THE GAMMA RAY TRACE Since these gamma-ray-absorption measurements, after being conducted in a laboratory batch mixer under controlled conditions, were also applied in the field while large concrete batches were produced continuously, each stage will be discussed separately. Laboratory Tests As described previously, while the mixing operation was in progress, changes in the physical characteristics of the absorbing media were continuously exposed to a collimated beam of gamma rays. These changes were converted into electrical signals, and then detected and amplified to produce a trace or continuous record of the variable which was changing with time. A typical gamma-ray trace obtained in the present investigations is shown schematically in Fig. 15. At Point 1, all the dry ingredients were charged into the mixer and were intercepting the radiation beam to reduce its intensity to appron‘mately 8000 counts per minute. The base line represents zero counts per minute. At Point 2, the weighed amount of water was added and consequently its effect was to increase the transmittance (Fig. 13). This increase in trans- mittance would indicate less bulk density than at Point 1. At Point 3, the curve not only rises markedly, but the amplitude of the needle swings increases greatly, both reaching a maximum at Point 4. At Point 5, the count rate and amplitude have both returned to about where they were at Point 3. Point 4 is invariably closer to Point 3 than to Point 5. Now the amplitude continues to decrease gradually and steadily until the end of the mixing period. A series -40- TRANSMITTANCE (COUNTS PER MINUTE) -41- Iqooo - s,ooo MIXING TIME (SECONDS) Figure 15. Schematic of a typical gamma-ray trace. 42- of 25 traces were obtained under similar. mixing conditions according to those procedures described before for laboratory experiments. At least two main features deserve particular attention. First, the remark- able manner in which the traces resemble each other almost exactly. They all follow the same pattern and they all Show at least a full minute of mixing before the fluctuations recorded on the trace smooth down. Second, the temporary rise in count rate between Points 3 and 5 which has been tentatively ascribed to that period of mechanical clumping. As explained previously, the mix water before being thoroughly distributed forms local pockets of low density between discrete "clods" of somewhat higher density. Obviously, if before mixing is complete, large portions containing more than their designed quantities of cement, water, gravel, sand, or air, should pass through the beam, their effect would be to cause an increase in transmission due to lesser average bulk density of the mixture (Fig. 13). Three of the 25 gamma ray traces obtained in the laboratory are shown in Fig. 16. Field Tests The field concrete plant consisted of three compartments with automatic scales to weigh separately the required amount of sand, gravel, and cement, prior to charging the large-size stationary mixer. The dry ingredients were carried on a belt-conveyor and discharged into the mixer. The duration of charging was about 20 seconds when the automatic equipment was operating tinder normal conditions. Water was weighed separately and gradually sprayed over the dry materials in the drum. After the prOper mixing time had elapsed, .meoeb mean-333w 3393. .3 93m?- -nozouunvuz.k oz_x_. o! On. ON. 0: oo. oo oo 2. 3 on 3 on ow C o. o L _ _ _ _ Exit/r13 (3.2.3.5471!) 2 m 2 m _ 2 _ 2 2 H 2 _ L _ fl §1§§§ .- .>. 5.233732.- 22, 223-3. 2.? . sign/(<3; ._. 2 2 _ 2 2 2 _ a 2 _ _ . max.) Gnu...- égg :5 E - C 57.3.} 53>- 532.222.122.232... $2: . 2} 3 c >k.0(a(u 0) DU 0 ~ # I — - fl _ _ - a n o 2 O U u n C u 2 - C. o z - x _ a on. .3- n: .3- 3 no . 2. no on at an 3 o. a a o! 02 ON_ o: oo. oo oo 2. on on at on ON 0. o o- _ _ _ _ _ _ _ ._ L _ b _ _ _ C — p _ _ _ _ _ _ _ _ — _ l— - ...r))-245LI-2II;¢«§PIali>ézs2e 33ch .3 mgswfl .898on marmamnomfic .552 .5 939m 450.33 P» so n 5; .358 a Bob 33.5 kahuna» Born .2 0.39,.” HoozouunHuIHh oszHs. % F b H :?H H H H HHH HHH H. HHHH. HHHHHHHHHH H. H H H H H\ H H H ,7 .HH 5, H, 5H. H22 6...; H5 .. HHHHHHHHHHHHHH H H. me x032 .ncmm MIXING TIME STUDY WITH SODIUM CHLORIDE Using standard laboratory equipment and mix materials, at leaSt 1 min of mixing is required to assure workable and durable concrete, according to conventional tests. However, because the photographic study just described indicated that workable concrete could be produced in a shorter time, further investigation was undertaken. In addition to the regular concrete ingredients which have already been mentioned, 200 g of dry sodium chloride was added. Then adding the required amount of water, the mixing operation proceeded as usual, with fourteen samples per batch taken at selected time intervals from a fixed point in the mixer. The rate of hardening of the fresh concrete was retarded by dis- solving 50 g of sugar in the mixing water before it was added to the mixture. In this way, each sample was weighed and then easily diluted in 2500 ml of distilled water. After filtering the mother liquor, 20 ml of solution was pipated and titrated with 0. 098 N silver nitrate according to a standard analytical procedure (47). The actual average value of the 200 g of sodium chloride added to the concrete batch was equivalent to 4. 8 ml of 0. 098 N silver nitrate per 1000 gm of concrete sample. The results from three concrete batches are shown in Table 5 and plotted in Fig. 21. In comparing these results with those obtained in the gamma ray traces, certain definite analogies were observed: 1. Initially, the silver nitrate-titration values scatter widely, approaching a limiting average value of 4. 80 ml of silver nitrate. After 75 seconds of mixing, the diSpersion is greatly reduced, falling within an experimental error -50- -51- ONN 53.820 Saleem as; beam was mass: .3 8sz 302033 ”.2; 02:22 on. 0! on. 8. oo on on ea a _-__4_1Aiij_,,n __, _ r r _~.._— . ._.- 0—,,_ ,rfix _ .L w_nr_m __ _._ _ =\\ :\.\ Lil \ "I‘ ./L . 3.34.3 420m» 4.? “TE ow 3.391.. 3