 From the moment a radioisotope shipment is received, it is handled with great respect for the potentially dangerous effects of its radiation. Even before any actual experiment has begun, the radioactive material undergoes repeated measurement. Some measuring instruments are used for personnel safety, so that the handler may know how much radiation is reaching his body and how long he can work at any given distance. The cutie pie and similar instruments determine the rate of radiation dosage. The total radiation dosage received by the individual worker in a given period, such as a day or a week, is measured by the film badge and the pocket ionization chamber, both worn wherever radiation is a possible hazard. Other instruments are designed for the scientific measurement of radioactivity. Those most generally used are the electroscope, the ionization chamber, the proportional counter, and the geiger counter. The operation of all these instruments is based upon the ionizing property of nuclear radiations. Nuclear radiations, alpha, beta, and gamma, in passing through matter, dissipate energy by producing ion pairs. When air or gas between oppositely charged electrodes is ionized, the positive ions are drawn to the negative electrode, and the negative ions are drawn to the positive electrode. If an auxiliary circuit is added, this ionization results in a flow of electrons. This is the principle of most electrical devices which detect and measure nuclear radiations. When the voltage on the electrodes is slight, say of the order of a few volts, most of the ion pairs recombine. Only a small fraction of the ions reach the electrodes. At this voltage, the number of ions collected is roughly proportional to the number and energy of the incident particles. However, in operation, as the voltage is increased, ion collection also increases until all the ions formed in the chamber are collected. The chamber is then said to be at saturation voltage. This illustrates the principle of the ionization chamber, the simplest type of radiation measuring instrument employing the ion collection principle. One type of ionization chamber is the electroscope. It measures ionization by the cancellation of a charge, seen by the movement of a leaf or fiber. Most ionization chambers have a sensitive current meter to measure the current in the circuit. Attainment of the saturation voltage is shown by a leveling off of the measured ion current as the voltage is increased. Other measuring instruments employ the ion collection principle at higher voltages. At increased voltage, the chamber enters a new phase of performance, illustrated by a single beta particle. At this voltage, the primary ions formed by the beta particle in the chamber are so accelerated in traveling to the electrode that they are themselves capable of ionizing the atoms of gas in the chamber. These secondary ions contribute to a further increase in ion current. Each particle is detected as a single current pulse. When operating at voltages in this range, the production of secondary ion pairs is localized to the region of the tube where the primary ions are formed. The production of secondary ion pairs increases with increased voltage. Each incident particle triggers a pulse of current in the circuit. The size of each pulse is proportional to the voltage and the energy of the incident particle, or the number of primary ions produced. This is schematically the operating principle of the proportional counter. If the voltage is greatly increased to the order of hundreds of volts, each secondary electron is accelerated, in turn producing a new generation of ions, until successive generations cause an avalanche of ions throughout the entire volume of the tube. Each incident particle of radiation initiates an avalanche of ions. This avalanche is observable as a single massive pulse of ion current. The total number of ions produced is the same for any radiation particle entering the sensitive region of the tube, regardless of its energy. This is the Geiger counter principle, in which each individual particle is detected as a single pulse of current. To stop or quench the discharge after each pulse, an auxiliary electronic circuit is used, or a special quenching gas mixture is employed in the tube. The period of time during which quenching occurs is called the dead time. During this period, another pulse will not register. A simple graph can be used to show the effect of voltage on the number of ions collected for each particle entering the tube. In the ionization chamber, only primary ions are formed. At saturation voltage, it collects all of the primary ions. In the proportional counter voltage range, the number of ions formed in the chamber is increased by the formation of secondary ions. Ion current is therefore proportional to the voltage and also the energy of the incident particle. In the Geiger counter voltage range, each individual incident particle triggers an avalanche of ions. The size of the resulting pulse depends on the voltage and is entirely independent of the energy of the incident particle. Even a low energy particle capable of producing only a few primary ions will cause the tube to register a pulse. The Geiger counter is the instrument most frequently used for radioisotope work. Although it is a versatile laboratory tool, it does not entirely replace the other measuring instruments. The electroscope is simple, inexpensive and maintains its calibration for years, but it is slow and not self-recording. The ionization chamber and its associated measuring equipment is used for specialized purposes such as the measurement of C14 labeled carbon dioxide and other radioactive gases. It is not used as extensively as the Geiger counter because of its relatively high cost and more difficult adjustment. The proportional counter is usually preferred for measuring alpha and low energy beta emitters such as carbon 14 and sulfur 35. The Geiger counter is in wide use because of its versatility in measuring many types of samples. The Geiger counter has certain operating characteristics which must be taken into account. Geiger counter tubes vary in design and therefore in operating characteristics. Each tube has an individual minimum operating voltage, known as starting voltage, below which it will not function. As the voltage is increased above the minimum, the instrument begins to record some of the impulses. Full efficiency is not attained until the so-called threshold voltage has been reached. Above threshold voltage, the efficiency of its measurement is constant over a considerable voltage range. This is called the plateau region. The investigator usually makes a graph of the number of pulses per minute or counting rate against voltage to determine the operating characteristics of any specific Geiger counter tube. Every Geiger counter has its own characteristic threshold voltage in plateau region. Beyond the plateau, a continuous discharge takes place inside the tube which may damage it. Therefore, for accurate measurement, the scientist must maintain the operating voltage above the threshold voltage and within the plateau region. In a Geiger tube, the occurrence of ionizing events can be seen by the pulsation of the meter. A mechanical recorder, actuated by the current pulses, is used to record the number of events. Such a recorder, however, requires larger pulses to actuate it than are generated in the tube. Therefore, a pulse amplifier must be used. Another characteristic of the recorder is its inability to register more than a few counts per second. In order to provide for the counting of samples of high activity, a scalar is introduced into the circuit. This scalar is an electronic device which passes on to the mechanical counter only one pulse for a certain number of pulses received from the tube. The proportion registered to those received is called the scaling factor. The scalar may have a factor of 2, 4, 8 and so forth, or it may have a factor such as 10, 100 or 1,000. The total number of counts recorded by the instrument is determined by multiplying the reading of the register by the scaling factor, which is commonly 64. To this total is added the number of counts which have not yet reached the scaling total. These leftover counts are obtained from the scaler's interpolation lights. In this case, 1, 2 and 16, or a total of 19. The result is the number of pulses measured in a specific interval of time by the scaler and recorder in the Geiger system. In radioisotope measurement, statistical data are always presented in terms of counts per unit time. Yet because of the random occurrence of individual decay events, the counts occur in a random manner. How accurate are the statistics? Suppose for instance that a 10 minute count has resulted in 1,000 pulses recorded, or 100 counts per minute. Yet the counts in any one minute vary considerably from the average because of the random occurrence of individual decay events. This gives an average for the 10 minute period of 100 counts per minute. There is a definite deviation in any one minute from the average for 10 minutes. Similarly, the 10 minute average also departs, although not as greatly, from the value which would be derived from a much larger total count. How accurate is the 10 minute measurement? In practice, one depends upon a concept of acceptable deviation by qualifying the total count by a plus or minus figure. A convenient acceptable deviation which is widely used is the square root of the total number of counts taken. For a large number of counts, this approaches what is known as the standard deviation. On this basis, with the total count of 1,000, one might expect a deviation in total counts of plus or minus 32, or a percentage deviation of 3.2%. This is a simple concept which can easily be explained by referring to the curve of statistical probability. Statisticians know from experience that in a series of measurements of a random process, most readings will fall near the average value, with relatively slight deviation. Some readings have a larger deviation, while a few vary greatly from the average. It is known that in a series of random events, two-thirds of the readings fall within plus or minus the square root of the number of counts. Thus, in using a deviation of the square root of the total count, 1,000 in this case, the statistician merely states that if the same measurement were repeated a number of times, the results could be expected to fall within a deviation of plus or minus 32, two-thirds of the time. This limit of deviation is one of several arbitrary conventions for defining the accuracy of experimental results. Taking the square root of the total counts, the deviations are respectively 10, 32, and 100 counts. But although the amount of deviation increases, the percentage of deviation decreases. In other words, one can expect the accuracy to increase with the square root of the number of counts taken. Where a standard deviation of 10% is acceptable, a total count of 100 would suffice. However, where great accuracy is necessary, 10,000 counts or more might be desirable. In practice, the Geiger counter instrument must first be warmed up at low voltage. Only then can high voltage be applied to the counter tube. Now the investigator must find the proper high voltage setting at which to operate the counter. The correct operating voltage varies from one tube to another. To determine the operating voltage, the investigator places a suitable sample of radioactive material below the counter tube. Next, the counter voltage is set at a level somewhat above threshold. At this voltage, a reading is made. This count is the first of a series of readings of the same sample at successively higher voltages. The first count is plotted on a graph of counts per minute against electrode voltage. Successive readings at increased voltages yield a graph which shows the operating characteristics of the instrument over a range of several hundred volts. On this curve, there is a nearly level plateau over which the counts do not vary appreciably with the voltage applied. For most work, the maximum permissible slope of the plateau is about 5% for 100 volts. Since tube life can be prolonged by employing the lowest practical high voltage, the operating voltage selected is low on the plateau, but far enough from any large irregularity in the curve to assure a fairly constant counting rate, even with slight voltage fluctuations. This selected voltage setting must be maintained throughout the experiment. A background determination is next made. For even in the absence of a sample, cosmic rays and natural radioactivity in the counting room cause an appreciable number of counts to be registered. Since the background counting rate is low, a fairly long counting time is required. For accurate measurement, its statistical deviation is usually taken into account. Background radiation will remain reasonably constant during a single experiment. The background counting rate must be subtracted from all further readings in order to correct the experimental results for these extraneous counts. Since background radiation does not vary appreciably over short periods of time, usually only one value needs to be determined for any one experiment. Radioisotope measurements, generally speaking, are of two types. In most measurements, the activity of one sample is compared with another. However, some experiments require exact knowledge of the rate at which atoms in the sample are decaying. This type of work is known as absolute measurement. To perform an absolute measurement of activity, one must take into account several factors relating the response of the instrument to the activity of the sample. The first factor is the geometrical arrangement of the sample and counter. Since the counter window subtends only a small angle at the sample, a correspondingly small fraction of the total radiation is directed toward the window. This angle changes with the position of the sample in the counting chamber. The next factor to be considered is the absorption of radiation between the sample and the counting region. Of all the radiation directed toward the window, some is absorbed by the cellophane which is often used to cover the sample. Some radiation is absorbed by the intervening air. Some is absorbed by the mica or glass of the counter window itself. Thus, the absorption of the radiation by cellophane, air, and window is a second factor which must be taken into account in an absolute measurement. The third factor is self-absorption of radiation within the sample. It is impossible to make an absolute measurement with accuracy if the sample is so thick that an appreciable amount of the radiation is absorbed within it and does not escape to be counted. The fourth factor is backscattering of radiation from the material on which the sample is mounted. With supports that are thick or of high atomic number, backscattering is increased. Thin materials of low atomic number such as plastic or aluminum keep backscattering to a minimum. A fifth factor is side-scattering of radiation from the interior of the sample chamber. This factor is small and will remain the same for similar samples placed in the same position. The total correction for all of these factors is determined by counting a standardized sample, that is one for which the absolute activity is known. A sample calibrated by the Bureau of Standards or another laboratory is used. The standard, a beta emitter in this case, serves as a base of reference. Of the several shelves, only one is ideal for a given sample. The investigator must first select the proper shelf. If the sample is too high in the chamber, the counter is not able to resolve individual particles. Some particles arrive during the dead time and are not registered. This loss of counts, by coincidence as it is called, is appreciable only at high counting rates. If the sample is too far from the counter window, the measurement will be slow and impractical. The ideal shelf is one which does not exceed the capacity of the equipment and yet will yield a valid count in a practical time period. In this way, coincidence losses are minimized without sacrificing the experimenter's time. Once the proper counting shelf has been selected, the investigator is ready to make a count. This count is made over a time period long enough to produce a statistically valid number of counts. To find the counting rate, the total counts are divided by the time elapsed. The previously determined background rate is then subtracted. This figure represents the rate at which beta particles from the standard sample are recorded by the Geiger tube. However, the fraction of the particles absorbed by cellophane, air, and window is not yet known by the investigator. Usually, a trained scientist carries out the correction for these absorptions since it involves a fairly complex procedure. In order to evaluate absorption by cellophane, air, and window, the investigator starts a series of measurements through aluminum filters of successively greater thicknesses. The density thickness of each filter is known in terms of milligrams per square centimeter. After counts have been taken with each of the filters, the resulting figures are corrected for background. A semi-log scale is used to plot the net counting rates against the density thickness of the filters in milligrams per square centimeter. The resulting graph shows the degree to which the radiation from the sample is absorbed by matter. On a semi-logarithmic scale, most of the curve is approximately straight. The point at which the curve meets the ordinate represents the counting rate without any filter in the counter. However, this point does not yet represent the counting rate with zero absorption of the radiation. It is necessary to extrapolate back past the ordinate to allow for the cellophane cover on the sample, whose density thickness in milligrams per square centimeter is known. An additional allowance must be made for the density thickness of the air between the sample and the tube window, in terms of milligrams per square centimeter. A further allowance is made for the window. Its density thickness is specified by the manufacturer. The result is the counting rate for the standardized sample, corrected for absorption by cellophane, air, and window. After extrapolating, the investigator knows the amount of radiation from the standardized sample with zero absorption as recorded by this particular counting arrangement. Calling this corrected counting rate R sub C, then let A sub S represent the absolute activity of the standard. Since the calibrated standard sample is used, the investigator knows the total beta activity, or the number of beta particles emitted per unit time by the standard in all directions. By dividing the corrected counting rate of the sample, R sub C, by the absolute activity A sub S, a geometry factor G is obtained. It represents that fraction of the total number of beta particles emitted by the sample, which would be measured by the counter, without absorption, but including scattering. The geometry factor will be the same for this counting arrangement when any beta emitter is being counted. In this experiment, the investigator is found that the geometry factor is 8.2%. In other words, a value of 82 counts will represent 1,000 particles emitted by the atoms of the standard sample. Having determined the geometry factor with the calibrated standard sample, the investigator is ready to determine the absolute activity of the uncalibrated experimental sample. The preparation of a counting sample usually requires chemical manipulation of potentially dangerous radioactive materials. For this reason, personnel safety procedures are followed, which have been developed over a period of years in radioisotope work. All critical operations are carried out in a fume hood, so that no contaminated air will escape into the laboratory. After dilution, the experimental sample is applied to a support of the same material and of the same thickness as that used with the calibrated standard. Thorough precautions are taken to prevent contamination of the investigator. To avoid ingestion, mouth suction is never used. A rubber bulb, glass syringe, mechanical pipetter, or aspirator is employed. The sample is pipetted onto the sample mount to produce the same thickness and area as with the standard. An infrared lamp is frequently used for drying samples. When the sample is completely dry, the investigator covers it with a thin sheet of cellophane held in position by cellophane tape to form a tight seal. This will avoid loss of the sample material, possible contamination of the laboratory, and inhalation or ingestion by laboratory personnel. To prevent contamination from being carried to the counting room where it could interfere with the accuracy of later measurements, the investigator's smock should not be worn outside the chemical handling room. Back in the counting room, the investigator starts the last phase of the measurement. To keep the geometry factor constant, the sample is installed on the same shelf which held the calibrated standard, a count is then taken of the sample. This count will represent the radiation reaching the counting area of the tube after passing through the cellophane cover, the intervening air, and the counter window. If the sample were the same radioisotope as the standard, the correction for absorption would be the same for both. But since the uncalibrated sample is not the same radioisotope, the energy of the beta particles emitted is different. For this reason, another series of measurements must be made through several graduated aluminum filters in order to find the proper correction for a different degree of absorption. As with the calibrated standard, the counting rates for the uncalibrated sample are plotted against the corresponding filter thicknesses. The curve is then extrapolated to correct for absorption by the cellophane cover, the air, and the window material. Once the corrected counting rate is known for the sample, it is only necessary to apply the previously determined geometry factor G to find the absolute activity of the sample. Thus, he has determined the total number of beta particles emitted per unit time by the radioactive atoms in the sample material. Absolute measurement of the rate of decay is basic to experimentation involving radioisotopes. It is primary and essential to the accurate evaluation of research in the isotope field. Frequently, as in tracer experiments, all that is required is that the activity of one sample be compared to another. In this comparative measurement, the activity of a sample extracted from the system under investigation is compared to the activity of a control sample to learn the quantitative change of activity produced by decay or dilution. Both samples must be mounted in the same position. In addition, the two samples must be prepared in the same manner. Also, the count must be sufficiently long to give the statistical accuracy required. Self-absorption can seriously affect the counting result. Exact comparison is best carried out with samples of identical thickness. A thin sample should be so thin as to eliminate any appreciable self-absorption. A thick sample should be infinitely thick, so thick that any greater thickness will cause no increase in radiation emitted. To determine the self-absorption of a sample of intermediate thickness, an involved procedure must be carried out. A series of samples of different thickness and known activity must be measured. Finally, for valid comparison, the sample's support should be identical to equalize the factor of backscattering. Illustrative of the simplicity of comparative measurement is an experiment described by Hevesi and Hahn, who used radioactive phosphorus to study the fate of blood phospholipids. The experiment is typical of most medical and biochemical tracer research. As frequently occurs, an animal is employed as a living laboratory to produce a needed reagent. The rabbit has received an injection of radioactive phosphorus. The blood sample, which is now removed, contains phosphorus-labeled phospholipids formed in the rabbit's bloodstream. Since the phospholipids are components of the plasma, the blood solids must be separated and the plasma isolated. For this purpose, a standard clinical centrifuge is adequate. After several minutes, the separation is complete and the supernatant plasma can be decanted. The scientist then sets aside part of the decanted plasma as control material. The remainder of the labeled phospholipids will be injected into a second test animal where it will be metabolized along with the animal's normal phospholipid supply. Comparison of the control material and the experimental sample will allow determination of the rate at which the phospholipids are metabolized in the test animal system. When the tagged phospholipids have been injected into the bloodstream and are permitted to circulate for a definite period, a blood sample is withdrawn. It contains some unknown proportion of the phosphorus atoms originally injected. The question is, how fast has the phosphorus been taken up from the rabbit's bloodstream? After this blood sample has been centrifuged, it is prepared for counting. The size and thickness of the evaporated sample are not important, since the control sample will be prepared identically. In order to serve as a basis for comparison, the control material from the first animal is diluted with normal rabbit plasma, equivalent in volume to that in the test animal. Since the mechanical dilution of the control sample is the same as that of the experimental sample in the animal system, any difference in their activities will be attributable to the rabbit's metabolic processes. After the control sample is measured for beta activity, it is removed from the counting chamber and the experimental sample is installed on the same shelf. By comparing the counting rates of the two samples, the scientist has determined the rate of movement of phospholipids out of the rabbit's bloodstream. This simple comparative measurement is valid only when the radioactive substance is of known identity and free of radioactive contaminants. In a case where the identity or purity of the sample is in doubt, its identity may be checked by determining its half-life. This is practical with isotopes of half-lives up to several months and is in itself a comparative measurement. In a half-life determination, a series of counts is made after several time intervals on a sample whose half-life and identity are either unknown or in doubt. Immediately thereafter, a similar measurement is made on a reference sample, which has a half-life so long that its decay can be ignored. This reference sample count permits the investigator to correct his results for any variations in the performance of the instrument between successive time interval measurements. After a series of measurements of the unknown sample has been made and corrected for counter-operation, the investigator is ready to plot the results. He plots a decay curve, a semi-log plot of activity against time. On the semi-logarithmic scale, the exponential decay of the unknown sample will be represented as a straight line. On the completed graph, the half-life can be read directly from the abscissa. Half-life can be used to make a tentative identification. A half-life experimentally determined to be 12.6 hours indicates that the sample is probably potassium 42. However, in this particular case, the identification is not so simple. For readings taken over an extended counting period show a change in slope. Therefore, it is evident that the unknown sample is contaminated with one or more additional radioisotopes, each decaying with a different half-life. In this case, K42 is mixed with a radioactive contaminant of considerably longer half-life. The plot of the longer-lived contaminant is a straight line. It may be extrapolated back to give its half-life about 14 days. The curve for the long-lived contaminant may then be subtracted from the observed curve. The corrected curve may then be drawn for the short-lived component. Now, from the half-life figure of 14 days, a tentative identification of the long-lived contaminant can be made. The literature indicates the contaminant is phosphorus 32. In this case, and in others where the half-lives are nearly the same, or one is too long for practical measurement, chemical studies may be required. Having tentatively identified the two radioisotopes, they can be separated into two samples and further identified by their radiation properties. After counting the two samples, the investigator draws an absorption curve for each of them. The curve of the beta-emitting component is drawn, a semi-log plot of activity versus density thickness. By extrapolating the curve to intersect the abscissa, the investigator obtains a point which represents the maximum range of the beta radiation. From this value, approximate beta energy in MEV can be determined by using a graph of range versus energy. This figure can also be obtained by the application of a simple empirical formula. Since radiation energies are specific for a given isotope, the figure for maximum beta particle energy gives added weight to the previous identification of the contaminant by its half-life. One component of the original sample was tentatively identified by its half-life as K42 before it was chemically separated from the P32. This component, like Phosphorus 32, can be checked and identified further by taking an absorption curve and finding the energy of its beta radiation. The absorption curve has no cutoff point, but flattens out horizontally. This indicates the presence of gamma rays expected from K42. A gamma ray absorption curve would further identify the isotope. Therefore, a second series of absorption measurements is made. For gamma measurement, lead filters instead of aluminum are used because of their high density and absorption. Except for the difference in filters used, gamma absorption is measured exactly as is beta absorption, and the data plotted in the same way. From the completed curve, the investigator determines the half-value layer or half-thickness. By reference to a graph of energy versus half-thickness, he finds the approximate energy of the gamma rays. Thus, by a third method, he has confirmed the identity of the gamma emitting component of his sample. The procedure this investigator has followed illustrates several typical applications of comparative measurement. He has combined technical proficiency and theoretical knowledge and applied them with understanding and foresight to the very practical problems of measurement. These techniques are common to all fields of isotope work. In absolute beta measurement, one of the basic techniques of isotope work, the rate of decay or activity of an unknown sample is determined by comparing it with a calibrated sample whose absolute activity is known. To do so, one must first take into account all factors which might cause experimental error. Absolute measurement is essential to the interchange of exact scientific data. Further, this accuracy is of paramount importance when, for reasons of safety to a patient, radiation dosage must be rigidly controlled. However, in most radioisotope work, comparative measurement is employed, in which an experimental sample is compared with a reference control sample. In a simple comparative measurement, the experimental sample, after passage through a process or system, is compared to an aliquot of the original stock material. Since the decay characteristics of both samples are identical, any difference in activity can be ascribed to the experimental process under study. Or, in another type of comparative measurement, the investigator determines the half-life of an unknown sample by comparison with a reference sample whose activity for all practical purposes is constant. In all comparative measurements, reference and experimental samples must be prepared identically and measured under identical conditions. The correct procedure of any measurement, to a large extent, sets the design of the isotope experiment. Meticulous techniques of measurement, whether absolute or comparative, are a free requisite to valid quantitative results.