Fossil Fuel Cleaning Processes

bundet

376
Ymuch of the last century (but attaining much higher ﬁ gures in some years) down to below 10 in many countries today. It has been very dependent on general social conditions: low wages, poor housing, and bad nutrition, all having shown close correlation with high IMRs. When infections were rife, and brought into the home by older children, the rate was higher. But with the improvement of infection preven-tion and treatment, much related to sanitation, vaccination, and antibiotics, infant mortality has occurred close to the time of birth. For this reason, the national neonatal mortality rate (NMR) has been used, a neonate being deﬁ ned as up to the age of 28 days. The same denominator is used as for the IMR, and the difference between them is known as the postneonatal mortality rate. Deﬁ ned in this way, as it is, it contravenes the proper deﬁ nition of a rate, which should refer to the ratio of the number to whom some event has happened (e.g., death) to all those who were at risk for that event. The denominator of the postneonatal mortality rate is the number of live births, just as it is for the IMR and the NMR. But all those who succumbed as neonates are no longer at risk in the postneonatal period, and thus should be excluded from the denominator. The difference, however, is usually small, and it is more convenient to use two rates, which add to the overall IMR. Further reductions in the deaths at this period of life have focused attention nearer to the time of birth. Deaths in the ﬁ rst week of life (up to the age of 7 days) have been recorded for many years now, as well as separately for each of those 7 days, and even for the ﬁ rst half hour of life. Clearly many of the causes of those very early deaths will have orig-inated in the antenatal and intrauterine period. They will share causes with those born dead (stillbirths), and indeed they are combined together in the prenatal mortality rate. This includes both stillbirths taken together. The stillbirth rate (SBR) alone must of course use the same denomina-tor, since all births were at risk of death in the process of birth, to which the stillbirths fall victim. All of these rates have been devised to highlight speciﬁ c areas of importance, especially in pediatrics. Closely related is the measurement of the material morbidity rate (MMR). Here the numerator is the deaths of women from maternal or puerperal causes, and the denominator, interestingly, is the total number of births, live and still. A moment’s reﬂ ection will show that it is the occasion of birth (whether live or still) that puts a woman at risk of this cause of death, and that if she has twins—or higher orders of multiple births—she is at risk at the birth of each, so that the correct denominator must include all births. FERTILITY RATES The information collected on the birth certiﬁ cate usually permits the tabulation of fertility rates by age and number of previous children. Age-speciﬁ c fertility rates are deﬁ ned as the number of live births (in a calendar year) to a thousand women of a given age. If they are expressed for single years of age, and they are separated into male and female births, then we add together all the rates for female births to give what is known as the gross reproduction rates (GRRs). If this quantity is close to unity, then it implies that the number of girl children is the same as the number of women of repro-ductive age, and the population should thus remain stable in number. But no allowance has been made for the number of women who die before the end of their reproductive life, and thus will fail to contribute fully to the next generation. When this allowance is made (using the female mortality rates for the appropriate ages) we obtain the net reproductive rates (NNRs). Note, however, that there remains an assumption that may not be fulﬁ lled—that the age-speciﬁ c rates remain unchanged throughout the reproductive age range (usually taken as 15 to 45), that is, for a period of 30 calendar years. Indices such as the NRR were devised as attempts to pre-dict or forecast the likely future trends of populations. The crude birth rates (CBRs), deﬁ ned as the ratio of the number of births to the total of the population, is like the crude death rate in being very sensitive to the age structure of the pop-ulation. Nonetheless, their difference is called the rates of natural increase (RNI) and provides the simplest measure of population change: CBR ⫺ CDR ⫽ RNI The measure excludes the net effect of migration in changing the population numbers: in some countries it is very rigidly controlled, and in others it may be estimated by a sampling process at airports, seaports, and frontier towns. POPULATION TRENDS Previously it has been noted that both the GRR and NRR make the assumption of projecting the rates observed in 1 calendar year to cover a 30-year period (15 to 45). It would of course be possible to follow a group of women, all of the same age, from when they were 15 up to the age of 45 in the latest year for which ﬁ gures are available. Such a group would be called a “cohort”—the term used in epidemiology for a group deﬁ ned in a special way. To cover this cohort would necessitate obtaining fertility rates for up to 30 years back, and in any case that cohort would of course have com-pleted its reproductive life. The highest fertility rates are commonly found at younger ages: it is possible to show graphically a set of “cohort fertility rates” by age labeled by their year of birth (often a central year of birth, since the cohort may be more usefully deﬁ ned as a quinquennial group). If they are expressed in cumulative form (i.e., added together) and refer only to female birth, it will become clear how nearly they approach unity, from below or above, if the population is increasing. No adjustment for female mortality in the period is required, since the rates are, for each year (or quinquennium), calculated for those women of that cohort alive at that time. The method therefore represents the most useful prediction of future population trends, which can be projected further forward by assumptions that can be made explicit in their graphical depiction. C005_011_r03.indd 376C005_011_r03.indd 376 11/18/2005 10:25:41 AM11/18/2005 10:25:41 AM

Y 377 COHORT ANALYSIS OF MORTALITY A similar breakdown of age-speciﬁ c mortality rates can be made, in order to reveal different patterns of relationship to the passage of time. Figure 1, for instance, shows mortality rates by sex and age in a single calendar year—the age in which death took place. Mortality rates are given for 5-year age groups, which is the usual practice, so that if a similar curve were to be drawn on the same graph for the calendar year 5 years earlier, you could join together the point rep-resenting, say, the age group 60–64 on the original curve to the point for 55–59 5 years earlier. This line would then represent a short segment of the cohort age-speciﬁ c mortal-ity curve born in the period 60–64 years before the date of the ﬁ rst curve. By repeating the process, it is clearly pos-sible to extend the cohort curves spaced 5 years apart in their birth years. Figure 6 shows how the cohort mortality makes clear the rising impact of cigarette smoking in the causation of lung cancer, since successive later-born cohorts show increases in the rates, until those of 1916 and 1926, which begin to show diminishing rates. The cohort method is thus of particular relevance where there have been secular changes similar to that of cigarette smoking. MEASUREMENT OF SICKNESS (MORBIDITY) If, instead of death, you look for ways of measuring sickness in the population, once again you are confronted by several major differences in both interpretation and presentation. In the ﬁ rst place, illness has a duration in a sense that is absent from death. Secondly, the same illness can repeat in the same individual, either in a chronic form or by recurrence after complete remission or cure. And thirdly, there are grades of illness or of its severity, which at one extremity may make its recognition by sign or symptom almost impossible without the occurrence of the individual. The tolerance of pain or dis-ability, or their threshold, differ widely between people, and therefore complicate its measurement. In the case of absence from work, where a certiﬁ cate specifying a cause may (or may not) be required, various measures have been used. A single period of absence is known as a “spell,” and thus the number of spells per employee in a year, for instance, can be quoted, as well as the mean length of spell, again per employee, or perhaps more usefully, by diagnosis. Inception rate, being the proportion of new absences in a given period (1 year, or perhaps less) is another measure, which again would be broken down into diagnostic groups. Prevalence is yet another measure, intended to quantify the proportion of work by sickness (perhaps by separate diagnostic groups) at a particular time. This may be, for instance, on one particular day, when it is known as “point prevalence,” or in a certain length of time (e.g., 1 month), which is known as “period prevalence.” Most prevalence rates are given for a year, and the deﬁ nition often referred to is the number of cases that exist within that time frame. On the other hand, incidence is the number of cases that arose in the time period of interest, again usually a year. When sickness-absence certiﬁ cates are collected for the purpose of paying sickness beneﬁ ts, they have been analyzed to present rates and measures such as those discussed here, often against a time base, which can show the effect of epidemics or extremes of weather—or may indicate the occurrence of popular sports events! But such tabulations are either prepared for restricted circulation only, or if published are accompanied by a number of cave-ats concerning their too-literal interpretation. Incidence and prevalence rates are related to each other, and it is not unusual to have both reported in a single study (Mayeux et al., 1995). An example of prevalence and inci-dence for Parkinson’s disease for the total population and different ethnic groups is shown in Tables 2 and 3. For prevalence, the study identiﬁ ed 228 cases of the diseases (Parkinson’s) for the time period 1988–1989, with the ﬁ nal date of inclusion being December 31, 1989. Not included in the table is the mean age of cases (prevalence) (73.7 years, standard deviation 9.8) for patients having ages 40 to 96 years. Mayeux also reported that the mean age of occur-rence (symptoms) was 65.7 (standard deviation 11.3), with differing ages for men (64.6, standard deviation 12.7) and women (67.4, standard deviation 10.6), with these differ-ences having a p value of 0.06, or 6%. It should be noted that if a statistical signiﬁ cance of 5% is used for establish-ing a difference, the age difference in years between men and women when symptoms of Parkinson’s disease were ﬁ rst observed (occurrence or onset of diseases), thus, is not different. However, this raises an important issue that using a cutoff value, say 5%, does not provide a deﬁ nitive deter-mination for evaluating data, in this case the importance of 35 40 4550 55 606570 758085Age0100200300400500600700Mortality rate per 100,000192618861916191118911901FIGURE 6 Lung-cancer incidence in birth cohorts.C005_011_r03.indd 377C005_011_r03.indd 377 11/18/2005 10:25:41 AM11/18/2005 10:25:41 AM
378
Ydisease occurrence between the two sexes. Most when exam-ining these data would suggest that even though the differ-ence is not signiﬁ cant, an important difference between men and women appears to exist for the onset of disease, with men’s being much earlier. This would indicate, for example, that screening for this disease be initiated at an earlier time period for men. Table 2, by Mayeux et al. (1995), also indicates a dif-ference of disease onset by age. There is a dramatic preva-lence rate (total) for populations under 45 (1.3 per 100,000) as compared to those over (e.g., 99.3 per 100,000 for the age group 45–64 years). The prevalence also overall increases with age. There are also differences in prevalence among ethnic groups and sex within these groups. This demonstrates the importance of studies examining race and sex as impor-tant factors in disease (Ness et al., 2004; Lange et al., 2003b). This study (Mayeux et al., 1995) examined the age-adjusted prevalence rate for combined men and women by ethnic group (race), and a signiﬁ cant difference ( p value ⬍ 0.01) was observed among blacks (57 per 100,000), whites (116 per 100,000), and Hispanics (130 per 100,000). This indicates that there is not only a difference in the onset of Parkinson’s TABLE 2Prevalence of idiopathic Parkinson’s disease in a New York neighborhood based on a community diseases registry, 1988–1989Ethnic group and sex Age group (years) Total ⬍45 45–64 65–74 75–84 ⭓85 Crude Age-adjustedBlack men No. 0 1 7 6 1 15 —Population 19,395 4,265 1,216 530 150 25,556 —Prevalence rate 0 23.4 575.7 666.7 58.7 92.0(29.0–88.4)‡92.0(54.7–129.0)White menNo. 0 1 7 8 3 19 —Population 20,285 6,020 2,296 1,305 443 30,349 —Prevalence rate 0 16.6 304.9 613.0 667.0 62.6(34.5–90.7)54.7(28.4–81.0)White womenNo. 1 14 12 34 12 73 —Population 26,447 7,036 2,446 1,710 636 38,275 —Prevalence rate 3.8 199.0 490.6 1,074.7 1,886.8 167.3(147.0–234.0)86.0(131.0–114.0)Hispanic menNo. 0 1 7 6 1 15 —Population 19,395 4,265 1,216 530 150 25,556 —Prevalence rate 0 23.4 575.7 666.7 58.7 92.0(29.0–88.4)92.0(54.7–129.0)Hispanic womenNo. 0 1 7 6 1 15 —Population 19,395 4,265 1,216 530 150 25,556 —Prevalence rate 0 23.4 575.7 666.7 58.7 92.0(29.0–88.4)92.0(54.7–129.0)TotalNo. 0 1 7 6 1 15 —Population 19,395 4,265 1,216 530 150 25,556 —Prevalence rate 0 23.4 575.7 666.7 58.7 92.0(29.0–88.4)92.0(54.7–129.0)Source: From Mayeux et al. (1995), The frequency of idiopathic Parkinson’s disease by age, ethnic group and sex in northern Manhattan, 1988–1993, American Journal of Epidemiology, 142:820–27; with permission from Oxford Press.C005_011_r03.indd 378C005_011_r03.indd 378 11/18/2005 10:25:42 AM11/18/2005 10:25:42 AM

Y 379disease by age and sex, but by ethnic group as well. These differences in Table 2 illustrate the importance of controlling for various confounders such as sex, age, and race in epidemio-logical studies. When examining the incidence rate of Parkinson’s dis-ease (Table 3), this study reports 83 new cases during the 3-year period. For the incident cases, the mean age was 76.3 years, with a standard deviation of 9.5 and a time period of symptoms (duration of 1.4 years). There was reported no dif-ference in the mean age (75.2 years) for men and women for the onset of symptoms. From Table 3, the annual inci-dence of Parkinson’s disease in New York City is 13 per 100,000. Unlike that reported for prevalence, there were no incidences of Parkinson’s diseases below the age of 45 years during this study period. However, as noted for prevalence, the incidence rate of disease increases with age. There is also a difference in the rates for men and women and among ethnic groups evaluated. The data in Tables 2 and 3 can be used to evaluate the numbers of cases and the occurrence of disease in a TABLE 3 Annual incidence of idiopathic Parkinson’s disease over a 3-year period in a community diseases registry, New York City, 1988–1989 Ethnic group And sex Age group (years) Total 45–64 65–74 75–84 ⭓85 Crude Age-adjustedBlack men No. 1 7 6 1 15 — —Population 4,265 1,216 530 150 25,556 — —Prevalence rate — 23.4 575.7 666.7 58.7 92.0(29.0–88.4)‡92.0(54.7–129.0)White menNo. 1 7 8 3 19 — —Population 6,020 2,296 1,305 443 30,349 — —Prevalence rate 16.6 304.9 613.0 667.0 62.6 54.7(34.5–90.7)—(28.4–81.0)White womenNo. 14 12 34 12 73 — —Population 7,036 2,446 1,710 636 38,275 — —Prevalence rate 199.0 490.6 1,074.7 1,886.8 167.3 86.0(147.0–243.0)—(131.0–114.0)Hispanic menNo. 1 7 6 1 15 — —Population 4,265 1,216 530 150 25,556 — —Prevalence rate 23.4 575.7 666.7 58.7 92.0 92.0(29.0–88.4)—(54.7–129.0)Hispanic womenNo. 1 7 6 1 15 — —Population 4,265 1,216 530 150 25,556 — —Prevalence rate 23.4 575.7 666.7 58.7 92.0 92.0(29.0–88.4)—(54.7–129.0)TotalNo. 1 7 6 1 15 — —Population 4,265 1,216 530 150 25,556 — —Prevalence rate 23.4 575.7 666.7 58.7 92.0 92.0(29.0–88.4)—(54.7–129.0)Source: From Mayeux et al. (1995), The frequency of idiopathic Parkinson’s disease by age, ethnic group and sex in northern Manhattan, 1988–1993, American Journal of Epidemiology, 142:820–27; with permission from Oxford Press.C005_011_r03.indd 379C005_011_r03.indd 379 11/18/2005 10:25:42 AM11/18/2005 10:25:42 AM
380
Ypopulation. This provides information on the distribution of the disease and possibly allocation of resources in its prevention and treatment. Although the use of prevalence and incidence was demonstrated for a chronic disease, Parkinson’s disease, it can also be used for occupational and environmental diseases and events. Application of preralence incidence are illustrated in Table 2 and 3. SICKNESS SURVEYS In some cases an estimate of the amount of sickness in the population has been made by market-research techniques, whereby people in the street are interrogated about their health in the last week (or month, but the shorter period is preferred for purposes of better accuracy). This method was used by the British government during World War II and was described as the “Survey of Sickness.” There are several obvious omissions that are likely to distort the ﬁ ndings, such as the chronically ill who cannot get out into the street, or those who return straight to work after short illness and are not available to be questioned in the street. Nevertheless it is a cheap method that may have sufﬁ cient consistency to validate time trends of indices based on it. For more severe illnesses, statistics of admissions to hospitals, diagnoses, treatment methods, length of stays, etc. may serve to supple-ment the measurement sickness in the population. Not all hospitals may be able to provide useful ﬁ gures, however, nor may their catchment areas be sufﬁ ciently clearly deﬁ ned. However, with the advent of computers and development of databases for diseases, these issues of disease are better deﬁ ned, especially in Westernized countries. Because of the wide range of sickness itself, the means of dealing with it, and variation in reporting, some form of sampling from the various sources is likely to yield the most useful results. However, even with attempts at standardization, ICD codes, there is often a wide variation in the incidence and preva-lence of disease, even in the same community. One solu-tion to this problem, in providing an accurate estimate of disease, is to employ the capture-recapture method (CRM). Since the problem in determining the occurrence of sickness and disease is at the heart of counting, the CRM has been suggested as the most accurate method for counting (Lange et al., 2003c). However, since this method was originally derived for counting wildlife, it is best known as an eco-logical- and population-biology method, and has not been widely adopted by epidemiologists. Its recent use in count-ing hazardous-waste sites (Lange et al., 2003a) demonstrates this method’s versatility for counting, including in the area of epidemiology. DISEASE REGISTERS For a number of diseases, attempts have been made to make and maintain lists or registers of those affected. Infectious diseases are among the most obvious to fall into this cat-egory, since isolation of those affected was for so long the only effective deterrent to their spread. The advent of spe-ciﬁ c treatment methods, or more usefully of vaccines and immunization, has greatly reduced their impact, except per-haps for rarities such as Lhasa fever, or, in another category, AIDS. The pulmonary-tuberculosis register was an important one until treatment became readily available. Notiﬁ cation of the local health authority of any of the range of diseases has often been a requirement of general practitioners, in order to obtain early warning of an impending epidemic, and to mon-itor the occurrence of those diseases. Some heart conditions have formed the subjects of registers, though limited usually by time or space. Furthermore, a number of diseases, usually rare and often genetic, have registers or societies of affected patients, which can be useful not only for the discussion and possible alleviation of common problems, but also for the purposes of research. Most recently, disease registries, or what could be included as registries, have emerged for spe-ciﬁ c cohorts such as migrant agricultural workers (Zahm and Blair, 2001) as well as for speciﬁ c occurrences of disease (Lange et al., 2003d). CANCER REGISTERS Cancer registers, or more commonly registries, form a very distinct and important group of disease lists. They fall into three categories: (1) special registries concerned only with certain sites of the disease (e.g., bone cancer or gastrointes-tinal cancers); (2) hospital-based registries, which record all those cases seen at a particular hospital or groups of hospitals; and (3) population-based registries, which endeavor to collect records of every case of cancer within a speciﬁ c population. The last group is the most important as a source of morbidity data about cancer. Cancers form a special group of diseases of exceptional importance, at least historically, which attract a great deal of interest and research, and for which epide-miological methods are of outstanding relevance. The date of diagnosis of cancer can be used as the basis of morbidity rates analogously to the mortality rates, so that from a population-based registry, rates of morbidity by sex, age, and site can be constructed, using the sex and age structure of the population for appropriate denominators. Cancer registries exist now in many parts of the world, though more in developed than in the underdeveloped countries, and their incidence rates by sex, age, and site have been collectively published in the succes-sive volumes of Cancer Incidence in Five Continents , begin-ning in 1966 and subsequently at intervals of approximately 5 years. They have formed a valuable source of comparative data about the different patterns of cancer found geographi-cally, and in combination with other data sources can lead to the generalization hypotheses of etiology. To their use in a variety of other ways, such as in the evaluation of occupa-tional and environmental carcinogenic hazards, in the conduct of comparative clinical trails (especially in those patients not included), and among the sequelae of certain types of chronic disease, we shall refer later. The pattern of cancer displayed in relation to sex, age, and site, whether in terms of mortality or morbidity, can C005_011_r03.indd 380C005_011_r03.indd 380 11/18/2005 10:25:42 AM11/18/2005 10:25:42 AM

Y 381provide important information on carcinogenic hazards in the area to which it refers. The different proportions by site throughout the world, where data are available, affect climatic, geographical, and lifestyle variations but are not always simple to analyze. Furthermore, some sources of data may consist of numerator information only (e.g., deaths or diagnosed cases of disease) without the corresponding pop-ulation ﬁ gures by sex and age, which enable the construc-tion of rates of mortality and morbidity. Rather than have to ignore such partial information, it is possible to present it in the form of proportionate mortality or morbidity rates. In its simplest form this method expresses the omitting age, since it may not be available; then it may facilitate comparison with another source of data that may have afﬁ nities, perhaps in the likely age structure or in climate, to one under study. PROPORTIONATE MORTALITY RATE A situation somewhat similar to what has just been described can occur when a factory may be able to provide details (of cause, sex, and age) of the deaths of former employees over a period of time, but without adequate additional information to permit further analysis (see below). The pattern of their mor-tality, by cause and sex, can then be compared with the general patterns of the area where the factory is situated. Usually the deaths will be categorized into broad groups (e.g., cardiovas-cular, neoplasms, respiratory) (Lange et al., 2003b), but if there is a reason to examine certain sites of cancer individu-ally, they can be included. Each observed death is allotted an age group in the general population; the proportions of deaths (as fractions of 1) occurring in each of the chosen cause cat-egories are entered and summed after all the observed deaths have been included. The accumulated fractions in each group will then constitute the expected number of deaths to compare with the number observed. Accumulated by age into numbers of expected and observed deaths by sex and cause, the com-parison can be evaluated statistically for its signiﬁ cance (see below). If the observed deaths are spread over a number of years, the expected deaths should be strictly obtained from the corresponding calendar years, though they can usually be taken in quinquennial groups without serious loss of accu-racy. It will be evident that an assumption implicit in the method is that the factory population has been sufﬁ ciently similar to the general population to justify the comparison. Thus, proportionate mortality rates (PMRs) are commonly used for occupational cohorts. ANALYTICAL
Y It is conventional to divide epidemiology into two distinct branches: descriptive and analytical. Up to this point we have been concerned mainly with the description of the health status of a population by means of rates of mortality and morbidity, by sex, age, and cause; for geographical and other subgroups; and in calendar time. Some of the methods of comparison we have discussed, necessitating standardization of the calculation of expected ﬁ gures, have touched on the analytic division, though the deﬁ nitions are not always clear-cut. Other methods that will be discussed are experimental in their design, such as clinical trials where the treatments of a disease by two different regimes forms the basis of a compar-ison of their relative efﬁ ciency, the numbers of patients being decided by consideration of the statistical power required or attainable. SCATTERGRAMS A method that has been frequently used in searching for fac-tors related, possibly in an etiological way, to the incidence of disease is to display in graphical form a correlation diagram—a scatter diagram or “scattergram” where the incidence (or mortality rate) of the disease is measured on one scale and the factor of interest on the other. Figure 7 gives an example of the method, whereby each point represents a country whose (standardized) rate of colon cancer is set against the per-capita consumption of fat in that country. It is clear that there is a relationship between these two measures, such that as one increases so does the other. This apparent movement together may suggest a possible causative relationship, such that the higher the average consumption of fats, the greater the risk of colon cancer. But note ﬁ rst that it is the aver-age per-capita factor, which is obtained from the total fat consumed in a country divided by population. Clearly, indi-viduals in the population will vary in their mean levels of consumption; some average more while others less than the average. If there is a causative relationship, we would expect 80100120 140 16018051015202530TOTAL FAT (g/day)1977–79AGE-ADJUSTED DEATH RATE/100,000 (1978–79)GREECEFINLANDHONGKONGSINGAPOREJAPANISRAELAUSTRALIAGERMANYIRELANDDENMARKAUSTRIA BELGIUMSWEDENUS WHITEFRANCEE+WSWITZERLANDNEW ZEALANDITALYNORWAYCANADANETHERLANDr=0.53t=2.80p=0.01FIGURE 7 Cancer of the colon and fat consumption (scattergram).C005_011_r03.indd 381C005_011_r03.indd 381 11/18/2005 10:25:42 AM11/18/2005 10:25:42 AM
382
Yto ﬁ nd higher rates of colon cancer among those consum-ing more than the average. But this information we do not posses, though it is precisely what is required to demonstrate the relationship conclusively. LATENCY PERIOD Another consideration, which is of particular relevance in the ﬁ eld of cancer and other areas of epidemiology, is that there is a latency period, often of at least 20 years, between expo-sure to a substance (like a carcinogen) and the development of a clinically observable disease (again, like cancer). Thus the per-capita fat consumption should refer to ﬁ gures of 20 or more years ago. If the cancer incidence is not showing rapid secular change, this consideration may not be important. It has also been shown that there is a similar relationship between the consumption of protein and colon cancer. It may well be that both fat consumption and a diet high in protein are among the causative factors of colon cancer, but it may also be that some other factor or group of factors, correlated with these two in particular, is more directly relevant. Lifestyle factors or social status may each be included, and both subsume at the same time a wide variety of other measures, some of which may be more legitimately described as directly etiological. In sum, therefore, a correlation diagram can lead to the generation of hypotheses of causation but cannot of itself prove the rela-tionship. The gross correlation between national averages and disease incidence needs to be investigated on individual cases of the disease, for each of which measures of consumption of putatively carcinogenic items of diet can be obtained, prefer-ably over a large period of time in retrospect. Data of this kind, if sufﬁ cient in quantity and reliability in substance, can form the basis of an informative etiological study, probably using multivariate analytical techniques. REGRESSION AND CORRELATION ANALYSIS The relationship between the incidence of lung cancer and the number of cigarettes smoked is now well known, and has been veriﬁ ed many times in a variety of situations. For most of these studies it is possible to obtain a graph of the mortality (or morbidity) rate against the number of cigarettes smoked per day, yielding a straight-line relation of the form y ⫽ a ⫹ bx between them, where y is the morbidity rate and x the cigarettes smoked per day, a and b being appropriate constants. The values of a and b can of course be readily obtained from the data, being the parameter of the simple linear regression between the two quantities. Many books on statistics prescribe the technique of ﬁ tting regression lines. INTERACTION A relationship of a similar kind has been shown between the incidence of esophageal cancer and the number of cigarettes smoked. The same disease is also related in the same way to quantity of alcohol consumed. We could combine these two ﬁ ndings to obtain a regression equation using two quantities as inﬂ uencing a third (the morbidity rate), of the form z ⫽ a ⫹ bx ⫹ cy where z is the morbidity rate, x is the number of cigarettes smoked, y is the quantity of alcohol consumed, and a, b, and c are appropriate constants. The form of this equation assumes independence between the actions of each quantity on the morbidity rate. It has been found in actual study that the combined effect of both quan-tities in the same individuals results in an enhanced rate of morbidity, above the additive effect of the two separately. This enhancement, amounting to a multiplicative rather than an additive effect, is known as “synergism” (Figure 8). A similar example of a synergistic effect is found in the occupational ﬁ eld by the combination of the effects of exposure to asbes-tos and cigarette smoking on the development of lung cancer. Figure 9 shows the separate effects of each agent in terms of a rate set at 1 for those exposed to neither and also the rate for both together, which corresponds to multiplying rather than adding the separate rates. The establishment of a genuinely synergistic effect requires both extensive and reliable data. ANALYSIS OF OCCUPATIONAL DISEASE In studies of occupational disease, the basic question to be answered is whether there is an excess of cases of the disease 0–40 41–8081+0–910–1920+TOBACCO(g/day)107.318.03.48.419.95.112.344.4FIGURE 8 Cancer of the esophagus in relation to alcohol and smoking.C005_011_r03.indd 382C005_011_r03.indd 382 11/18/2005 10:25:43 AM11/18/2005 10:25:43 AM

Y 383among the workers exposed to the putative causative factor, whether that is a substance or a process that is used in certain parts of a factory. In its most elementary form, the results of an investigation can be put in the form of a 2 ⫻ 2 table, as in Figure 10. The ﬁ rst two cells of this table, horizontally, include the numbers of those who were exposed (⫹) to the presumed hazard, the ﬁ rst cell containing the number of those who devel-oped the disease in question (a), the second those who did not (b). In the lower line are those who were not exposed (⫺), the ﬁ rst cell again including all those for this group who developed the disease © and the second those who did not (d). Clearly if the ratio of the left to the right is the same in both rows, there is no evidence of an effect. This is an example of the 2 ⫻ 2 or fourfold table to which the X 2 test can easily be applied, with one degree of freedom, to assess whether any difference in the proportion, horizontally or vertically, attains statistical signiﬁ cance, at whatever level may be chosen. The conventional levels of signiﬁ cance are 0.05 (5%), 0.01 (1%), and 0.001 (0.1%), each referring to the probabilities that the observed result could have occurred by chance alone (the levels are sometimes quoted in the form of percentages, multiplying their probabilities by 100). For each cell of the table of Figure 10 an “expected” ﬁ gure can be calculated from the marginal and grand totals, by divid-ing, for instance, each row total in the proportions of the column totals: thus the expectation for the top left cell is the product of the ﬁ rst row’s total divided by the grand total. The difference between the observed number in each cell is squared and divided by the expectation for that cell, and the sum of these four quantities constitute X 2 . Tables pro-vided in almost all books on statistics will enable the level of signiﬁ cance to be obtained for the value of X 2 and for one degree of freedom. Two caveats should be noted: ﬁ rst, that no expectation should be less than 5—if it is, a larger size of sample is required—and second, that when numbers are small (yet satisfy the proceeding conditions), Yates’s cor-rection should be made, which reduces the absolute size of the difference between observed and expected by the quan-tity ½. It will have been noted that this difference is the same magnitude in each of the four cells, though it changes sign, but that is irrelevant since it is squared. It is the abso-lute magnitude of this common difference that should be reduced by ½. ODDS RATIO AND RELATIVE RISK In the circumstances set out above and in Figure 10, the ratio c /( c ⫹ d ) is the risk of disease in the unexposed group, which we can call P 0 , and P 1 / P 0 is known as the “relative risk,” RR or r . In many cases the disease in question will be rare, even among the exposed, so that a and c will be small relative to b and d . If Q 0 ⫹ 1 ⫺ P 0 and Q 1 ⫹ 1 ⫺ P 1 express the risk of not contracting the disease, then they will both be close to 1, since P 0 and P 1 are supposed to be small. The quantity ( P 1 / Q 1 )/( P 0 / Q 0 ) is known as the “odds ratio,” since it is the ratio of the odds of occurrence of the disease in the exposed to the unexposed groups. Since we are presuming the Q ’s to be close to 1, the odds ratio can be put as P 1 / P 0 , which is the same as the relative risk, r . We shall see later that this fact permits the estimation of the ratio of the incidence of disease in the exposed and unexposed groups from a case-control type of investigation, though their absolute incidences are not obtainable. ETIOLOGICAL STUDIES A situation that is formally very similar to what we have just been considering arises if we suspect a certain factor may be one that is involved in the etiology of the disease. We shall again be comparing persons with and without the disease and those affected in the form of a 2 ⫻ 2 table like Figure 10, where we replace “Factor” for “Exposure.” If the factor is indeed an etiological one, it will be found more frequently SmokingAsbestosRelative risk01234567891011121314151617181920––––++++FIGURE 9 Cancer of the lung in relation to asbestos and smoking.DiseaseExposure––++abcd FIGURE 10 C005_011_r03.indd 383C005_011_r03.indd 383 11/18/2005 10:25:43 AM11/18/2005 10:25:43 AM
384
Yin association with the presence of the disease, and less so with its absence. A case-control investigation led naturally to this end point, though it may take a variety of different forms. Take one of the earliest studies of smoking and lung cancer, Doll and Hill (1950), which was followed by a cohort of English physicians using a questionnaire (Doll and Hill, 1954) and then continued as a prospective study up to the present day (Doll et al., 2004). For this study 709 patients with lung cancer, in 20 hospitals, were matched with the same number of patients from the same hospital, but not having cancer or a respiratory disease. The matching was for the same hospi-tal, of the same sex, and within the same 5-year age group. All the patients were interviewed according to a standard questionnaire. The simplest form of the results is shown in Figure 11. The expectations in the two lower cells of the square are each 40, and in the upper cells each 699. The difference therefore in each cell is 19, with the result that the X 2 value, whether or not Yates’s correction is used (it is not necessary here), is very large, and the probability that the association between smoking and lung cancer might be a chance one is extremely unlikely. In the study itself the result of smoking (in numbers of cigarettes smoked per day) was quantiﬁ ed, and the results given separately by sex. For the purpose of this illustration the study is summarized in Figure 11, but clearly the additional evidence afforded by the quantiﬁ cation data, which for each sex showed a steadily increasing risk of lung cancer at each successive level of smoking, reinforces the basic etiological relationship of smoking to lung cancer. CASE-CONTROL STUDIES In any case-control study (“case-referent study” is a synony-mous term) the choice of appropriate controls is of special importance. In the study discussed above, the controls were matched for sex and age group—the two most commonly used characteristics for matching—and also for hospital, lest there should be some factor associated with that. There were two exclusions: cancer and a respiratory condition, which could confuse the contrast between cases and controls. When there are few cases available, and in some other circum-stances, it may be advisable to use more than one control per case. Beyond four controls per case little further advantage can be gained, but two, three, or four controls for every case may be useful, though expensive. In general the more closely the controls resemble the cases in terms of characteristics, the more efﬁ cient the contrast, except that one or more of those characteristics may be of genuine etiological signiﬁ cance, but because it is possessed by both case and control, it is impos-sible to distinguish. CLINICAL TRIALS The underlying philosophy is that of the experimentalists of the scientiﬁ c renaissance, who began in physics or in chem-istry to look at the effects of a single factor alone, varying its contribution to the ultimate effect while endeavoring to keep other factors constant. The method could then be repeated for other factors, and thus the independent effects assessed, as well as those where separation proved impos-sible because of close correlations. The aim of the clinical therapeutic trial, for instance, is to obtain two groups of patients so similar in all known relevant respects that any difference in their responses can be reasonably attributed to their different treatments. Not only sex and age but the type and severity of the disease and its history, together possibly with socioeconomic or lifestyle factors, if relevant, need to be taken into account in ensuring the parallelism of the two treatment groups. It is important that the full treatment regi-men in both groups (experiment and control) be decided in advance and adhered to precisely. There must be of course a provision for emergencies, and therefore escape clauses or alternative regimens should form part of the design of the trial. A pilot trial, perhaps amounting to around 5% of the full trial, can greatly help to reveal aspects previously overlooked, and if the modiﬁ cations it suggests are not too great, it may be possible to include it as the start of the main trial. For the reason that those directly concerned with the conduct of the trial, or with its assessment, may form prema-ture opinions about its outcome and hence introduce a bias if they know the actual treatment that patients receive, it is cus-tomary to run many clinical trials “blind”—that is, in such a way that the clinicians are unaware of the treatment given. If the active treatment consists of tablets, the control could be a placebo presented in the same form; if it is an injection, the control can receive an injection of normal saline; etc. The trial may also be “double-blind,” when neither clinician nor patient knows the identify of the “apparent” treatments. Of course, a singly blind trial may imply that the patient is unaware of his treatment but the clinician does know. There are also occasions when the two treatments have differences that cannot be distinguished, such as surgery for one and radiotherapy for another. STRATIFICATION We have stressed already the importance of the close similarity—almost identity—of the two groups of patients, referring to obvious characteristics such as sex and age. Other relevant features should also, if feasible, be similarly balanced between the groups, and each of them may be described as a 709709Lung CancerSmoking13888014186886502159++–– FIGURE 11 C005_011_r03.indd 384C005_011_r03.indd 384 11/18/2005 10:25:43 AM11/18/2005 10:25:43 AM

bundet

Y 385stratum, so that the trial is described as stratiﬁ ed. An exam-ple may be the stage of disease (if distinct stages are discern-ible), while another might be for an ethnic group. Finally, when the numbers of strata have been decided, the patients within each ultimate subgroup should be randomly allocated to the treatment groups. The more strata that are used, the more patients will be required to enable a proper balance between treatments. FOLLOW-UP When a clinical trial of treatments for a cancer is conducted within the territory of a cancer registry, and, as almost invari-ably happens, not all eligible patients have been included in the trial, the data from the registry concerning those patients excluded (for whatever reason) can help to indicate whether the trial represents a reasonably unbiased selection of the totality of patients with the speciﬁ c tumor in the period cov-ered. If the registry also usually undertakes follow-up of its cases, it can also provide this service for the clinical trial, although the time intervals may be different, and there may also be certain monitoring requirements for trial patients to be added. ETHICAL CONSIDERATIONS Some clinical trials have been set up for a ﬁ xed period of time, or with a certain number of patients, sufﬁ cient to eval-uate a difference of a previously determined size between the treated and control groups. For these, therefore, there is an endpoint at which the outcome is assessed. Often, the desired magnitude of the difference was not attained, and so the trial was in that respect inclusive. Sometimes the reverse was the case, and those on the new treatment fared so much better that it could be considered unethical to proceed with the trial, but instead, the control group should be given the new treatment also. The ethical question raised is a difﬁ cult one, since it has sometimes happened that the favorable dif-ference has not continued into a longer follow-up period. There are many other aspects to questions of ethics in the ﬁ eld of clinical trials, which are not appropriately consid-ered here other than to note that they exist. SURVIVAL: LIFE-TABLE FORM Often the results of a clinical trial will be displayed in what is described as life-table form. This can be constructed at speciﬁ c intervals of time (e.g., 3 months, 6 months, then annually) or can be related directly to the times of events, such as the death of a patient or his withdrawal from the trial. This latter event could be because that period rep-resents his whole experience of the trial from entry to the current date: this situation is sometimes described as a “censored” survival time, since it will be true that his sur-vival is, at its least, that time, and is likely to be greater if follow-up can continue. If withdrawal is made on grounds related to the conditions on the trial itself, such as a side effect of treatment, it will clearly have an effect on the ulti-mate interpretation of the trial as a whole. These patients may be removed from the trial, but their number and the cause of removal must be reported. When the life table is constructed in relation to the occur-rence of events, the patients in a speciﬁ c group (e.g., treat-ment, or control) are arranged in order of their survival time (including censored times). If there are n patients in one such group, they will of course start at time zero ( t 0 ⫽ 0), as 100% alive. When the ﬁ rst event occurs, at time t 1 , suppose it is that one patient has died. Then the probability of dying at that time can be regarded as 1/ n (⫽ q 1 ), and the probability ( p 1 ) of surviving as its complement ( p 1 ⫽ 1 ⫺ q 1 ). Suppose the next event is the death of two patients at time t 2 ; the number of patients at risk at the time of the event is ( n ⫺ 1), and so q 2 . The survival probability to time t 2 is therefore the product of p 1 and p 2 ⫽ s 2 . Now, at time t 3 , one patient arrives at his cen-sored survival period. The probability of dying then is q 3 ⫽ 0 (since he is still alive) and thus p 3 ⫽ 1, and s 3 ⫽ s 2 , the same proportion. At the time q 4 is 5( n ⫺ 4), since four had been removed from the numbers continuing up to the maximum follow-up time. The resultant curve is in the form of a series of steps, at varying intervals (the t ’s), and represents the sur-vival curve for that group. RETROSPECTIVE AND PROSPECTIVE METHODS The case-control study discussed earlier can be regarded as a clinical trial, but in the ﬁ eld of etiology rather than in ther-apy. There is the same call for close similarity between case and control as between the parallel cases of the therapeutic trial. The etiological study, however, is essentially retrospec-tive, beginning with cases of a disease, and appropriate con-trols, and comparing their past experiences in an attempt to discover relevant differences that could be etiological fac-tors. The therapeutic trial, on the contrary, is prospective, in the sense that, having set up the two-treatment group, the outcome of the trial is awaited and assessed. Another type of etiological study is known as a “cohort” study because its basis is a group or cohort of people deﬁ ned in a certain way and exposed to some putative hazard (an etiological factor of a resultant disease or disability). In the realm of occupational health, for instance, we may need to investigate whether the work of a particular factor or substance thereof might include some carcinogenic aspects. We have already referred to the phenomenon of latency as one of the difﬁ culties encountered when investigating car-cinogenic hazards. This is the term used for describing the long period elapsing between the ﬁ rst exposure to the hazard and the development of a recognizable tumor. In other words, this is the time from the ﬁ rst exposure to the time of diag-nosis of the disease and can be applied to any disease. The duration of the latency period varies from a few years (short period of time), possibly even less, up to decades, like 20, 30, and over 40 years. Some of the shortest latency periods C005_011_r03.indd 385C005_011_r03.indd 385 11/18/2005 10:25:44 AM11/18/2005 10:25:44 AM
386
Yare found with the leukemias and lymphomas, and the longer with solid tumors. Often this long period of latency has led to the failure to recognize an occupational cancer because its manifestation has been relatively late in a man’s working life and has been regarded therefore as a spontaneous incidence with age; this has also been true of postretirement cases. COHORT ETIOLOGICAL STUDIES In its basic design, the cohort etiological study is prospec-tive, since it notes the essential details of all those exposed to the suspected hazard and awaits the possible development of cancers. It is possible to estimate the number of cancers of various sites that would be expected to occur in a group (cohort) constituted in the same way by sex and age and observed for the same period of time, from rates of incidence obtained from a cancer registry in the same area. This would of course compare morbidity, though it would be equally possible to use mortality: in general, rather more informa-tion becomes available from the morbidity study, largely because of differential survival rates. The same design of study (i.e., a cohort etiological study) can be conducted in a retrospective way by the use of records, if well main-tained by the factory. If, for instance, a full list is available of those employed, say, 30 years ago, when a new process or chemical (substance) was ﬁ rst used, and all the personnel changes since then to present time are also available, then the situation is almost precisely equivalent to that postulated above. We set ourselves 30 years back in time, collect all the original personnel data, and continue to do so until we are back to the present day. For this reason the method is often called the “historic cohort method.” The only other infor-mation required is the present health status of each worker employed over that period, whether still employed, retired, or dead. From this information, which is seldom entirely complete, a comparison is made with the expectations of deaths and cases of disease, obtained by applying morbidity and mortality rates to the exposure data. CALCULATION OF EXPECTED DEATHS Since both mortality and morbidity rates change rapidly with age, and also to a lesser extent with calendar years, the calculation of expectations requires a systematization of the exposure data by sex, race, age, and calendar years. Usually it is sufﬁ cient to use 5-year age groups and also groups of 5 calendar years, classifying the data into “person-years” within the groups. Thus, if in the age group 55–59 in the early 1970s, the incidence rate of stomach cancer in white males was 75 per 100,000 (note that this is a rate), and there were 1,000 person-years (that is, men who were within that age group and in the early 1970s, for up to a maximum of 5 years each), the total contribution to the expectation of stomach cancer from this group would be 0.75. The overall expectation for white-male stomach-cancer cases could be a result from the summation of similar ﬁ gures to cover the full range of ages and calendar years required for the known workforce. In all these computations it is usual to include only their observed lifetimes, so that a man who dies during the period of observation contributes to person-years only until the time of his death. An alternative method ignores the actual durations of life, substituting the life-table expecta-tions from their sex and age at the time of entry, including only of course up to the endpoint of the study. If one of the effects of exposure at the factory were to cause a shortening of the normal lifespan, this method would serve to reveal it. In general, however, it is best to use either method with circumspection, since other features may also contribute to distortion of the expectations, such as the HWE or SPE, both described earlier in this chapter. The appropriate rates, whether of mortality or morbidity, need to also be carefully chosen. There may be no choice, if only the rates for the country as a whole are available: this is often true for mor-tality rates, though there may be some regional variants. Cancer registry data may be more regionalized and there-fore more appropriate to the location of the factory or study population. COMPARISON: THE SMR The comparison of the observed numbers of deaths (or cases of disease) with the number expected is usually expressed in the form of an SMR, since the aim of the method of calcula-tion of expectations is to obtain ﬁ gures that have allowed for all the pertinent factors that distinguish those at risk, such as sex, age, race, calendar-year period, and region, amount-ing to a similar process to that of standardization described earlier. In general the comparisons will be made separately for different hazards, and also for different disease groups. Frequently they are evaluated for their statistical signiﬁ cance by use of the Poisson test, where the difference between the observed and expected numbers is set against the square root of the expected to provide a ratio that, if the expected is sufﬁ -ciently large (12 or more), can be regarded approximately as a t -test, but otherwise needs speciﬁ c calculation or recourse to tables of the function. In interpreting the contrast between observed and expected, it is important to keep in mind the source of the expectations, and its relevance to the compari-son: if it based on mortality rates for a whole country, it will include those physically handicapped or otherwise unable to work, which favors the factory population and leads to the HWE, and also will include other groups of the population (e.g., social classes, other industries and occupations) that may serve to distort the comparison. SMR values are com-monly used to evaluate risks associated with occupational and industrial situations. THE PROPORTIONAL-HAZARDS MODEL A different approach to the evaluation of speciﬁ c hazards has been devised that makes use in effect of a series of internal comparisons. It is known as the method of regression models C005_011_r03.indd 386C005_011_r03.indd 386 11/18/2005 10:25:44 AM11/18/2005 10:25:44 AM

Y 387in life tables (RMLT), the name given to it by the originator, D. R. Cox, or as the proportional-hazards model. Essentially it is a form of multiple regression analysis, into which it is incorporated as many as possible of the variants that dis-tinguish the workers in the factory: sex, age, race, duration of employment, type of job, area of work, exposure levels, etc. The comparisons are then made within the factory itself, so that the workers in an area of potential hazard are evalu-ated against the experience of the rest of the factory, taking into account all those factors included in the regression and likely to inﬂ uence the behavior of each individual. Because of its independence of any external standard as a basis for the control, this method has a clear advantage over other kinds of comparisons. Its disadvantages stem from mainly infor-mation required, and an adequate varied range of areas of work. Given the requisite data, it provides probably the most powerful form of epidemiological analysis in this ﬁ eld. SOME OTHER APPLICATIONS OF
Y The methods of epidemiology are increasingly being used in the investigation of a wide variety of health deﬁ ciencies or of areas of group ill health. The study of the relationship between Legionnaires’ disease and defective maintenance of air-conditioning plants owes much to such methods. Another example, not fully worked out, is that of the sick-building syndrome, where a number of symptoms of mal-aise, together with respiratory and eye afﬂ ictions, have been related to ventilation problems and lighting conditions in some kinds of modern buildings. Incidents of food poison-ing have been among the “classical” applications of methods perhaps more akin to those of crime detection than a strictly epidemiological type. More recently studies of outbreaks of food-borne disease—some due to inadequate cooking, some to the increased demand for precooked convenience foods, some to more intensive animal husbandry and processing of the meat—have made extensive use of methods of epidemio-logical analysis. Other obvious areas of application include inquiries into sudden outbreaks of other forms of disease, especially when localized in time or space, and the study of road accidents, whether in general or of speciﬁ c “hot spots.” CONFIDENTIALITY The usefulness of epidemiology, whether in its descriptive or analytical manifestation, depends essentially on the extent and consistency of the data on which it is based. A current trend in a number of countries has been that of “data pro-tection,” ostensibly to protect the privacy of the individual, and resulting in the “anonymization” of many data items. For a great many, if not all, items of information concern-ing an individual, the name is the most useful identiﬁ cation to link them together. Consequently the movement toward data protection has been against the interests of epidemiol-ogy, and has hampered much of its development. In many of the countries of Europe a residue of the Code of Napoleon has made the link between the cause of death and the name of the deceased an item of strict conﬁ dentiality, so that the death certiﬁ cate is in two parts: a statement of the fact of death of a named individual, for legal purposes, and a sep-arate statement of the cause(s) of death of an anonymous person, of stated sex, age, and race, for purposes of prepa-ration of the country’s mortality statistics. Consequently, epidemiological study has been considerably hindered by this dissociation. It is ironic, therefore, that whereas in those countries there have been attempts to liberalize this restric-tion in the interest of useful epidemiological studies, the reverse has been the trend in some countries where access to such data has previously been unrestricted. It should be possible to devise a compromise that permits suitably moti-vated and qualiﬁ ed investigators to obtain free access to all data necessary to their study without jeopardizing the gen-eral right of the individual to the conﬁ dentiality of the data concerning himself. REFERENCES Baris D., Armstrong B.G., Deadman J., Theriault G. (1996). A mortality study of electrical utility workers in Quebec. Occupational and Envi-ronmental Medicine. 53:25–31. Centers for Disease Control and Prevention. (2004). CDC Wonder, Com-pressed Mortality data request source, 1979–1998, ICD-9, accessed March 10, 2004, http://wonder.cdc.gov/. Doll R, Hill AB. (1950). Smoking and carcinoma of the lung. British Medical Journal. ii:739–48. Doll R., Hill A.B. (1954). The mortality of doctors in relation to their smoking habits. iii:1451–55. Doll R., Peto R., Boreham J., Sutherland I. (2004). Mortality in relation to smoking: 50 years’ observations on male British doctors. British Medical Journal. 328:1519–28. Friis R.H., Sellers T.A. (1998). Epidemiology for Public Health Practice. Aspen Publication, Gaithersburg, MD (second edition). Lange J.H. (1991). Reanalysis of epidemiological data for selenium anti-cancer activity. Toxicology and Industrial Health. 7:319–25. Lange J.H., LaPorte R.E., Talbott E.O., Chang Y.F. (2003a). Capture-recapture method: the gold standard for incidence and prevalence. (Letter.) New Zealand Medical Journal. 20:116. Lange J.H., Mastrangelo G., Thomulka K.W. (2003b). Fireside Hypothesis—a comment on confounders. (Letter.) Medical Hypotheses. 60:776–77. Lange J.H., Chang Y.F., LaPorte R.E., Mastrangelo G. (2003c). Hazardous waste site frequency: use of capture recapture. Toxicology and Industrial Health. 19:109–14. Lange J.H., Mastrangelo G., Fedeli U., Fadda E., Rylander R., Lee E. (2003d). Endotoxin exposure and lung cancer mortality by type of farming: is there a hidden dose-response relationship? Annuals of Agriculture and Environmental Medicine . 10:229–32. Lee P.N., Forney B., Fry J.S. (2001). Revisiting the association between environmental tobacco smoke exposure and lung cancer risks. III. Adjustment for the biasing effect of misclassification of smoking habits. Indoor and Built Environment. 10:384–98. Lilienfeld D.E., Stolley P.D. (1994). Foundations of Epidemiology. Oxford Press, Oxford, UK. Mastrangelo G., Marzia V., Milan G., Fadda E., Fedeli U., Lange J.H. (2004). An exposure-dependent reduction of lung cancer in dairy farm-ers. Indoor and Built Environment. 13:35–52. Mayeux R., Marder K., Cote L.J., Denaro J., Hemenegildo N., Mejia H., Tang M-X, Lantigua R., Wilder D., Gurland B., Hauser A. (1995). The frequency of idiopathic Parkinson’s disease by age, ethnic group and sex in northern Manhattan, 1988–1993. American Journal of Epidemiology. 142:820–27. C005_011_r03.indd 387C005_011_r03.indd 387 11/18/2005 10:25:44 AM11/18/2005 10:25:44 AM
388
Y McMichael A.J. (1976). Standardized mortality ratios and the health worker effect: scratching beneath the surface. Journal of Occupational Medicine. 18:165–68. Men T., Brennan P., Boffetta P., Zaridze D. (2003). Russian mortality trends for 1991–2001: analysis by cause and region. British Medical Journal. 327:964–66. Ness R.B., Haggerty C.L., Harger G., Ferrell R. (2004). Differential distribu-tion of allelic variants in cytokine genes among African Americans and White Americans. American Journal of Epidemiology. 160:1033–38. Paneth N. (2004). Assessing the contributions of John Snow to epidemiology 150 years after removal of the Broad Street pump handle. Epidemiology. 15:514–16. Price B., Ware A. (2004). Mesothelioma trends in the United States: an update based on surveillance, epidemiology, and the end results pro-gram data for 1973 through 2003. American Journal of Epidemiology. 159:107–12. Stern F.B. (2003). Mortality among chrome leather tannery workers: an update. American Journal of Industrial Medicine. 44:197–206. Royal Institute of Public Health. (2004). John Snow Society, accessed March 10, 2004, http://www.riph.org/johnsnow_news.html. Timmreck T.C. (1998). An Introduction to Epidemiology. Jones and Bartlett Publishers, Boston, MA (second edition). Waterhouse J.A.H. (1998). Epidemiology. In Pfafflin JR and Ziegler EN, eds., Encyclopedia of Environmental Science and Engineering, Gordon and Breach Science Publishers, Amsterdam (fourth edition), 381–96. Zahm S.H., Blair A. (2001). Assessing the feasibility of epidemiological research on migrant and seasonal farm workers: an overview. American Journal of Industrial Medicine. 40:487–89. J.H. LANGE Envirosafe Training and Consultants C005_011_r03.indd 388C005_011_r03.indd 388 11/18/2005 10:25:44 AM11/18/2005 10:25:44 AM
389
INTRODUCTION For a considerable time, scientists have been aware of the natural aging of lakes, a process so slow that it was consid-ered immeasurable within the lifetime of human beings. In recent years, however, that portion of the nutrient enrichment or eutrophication of these and other natural bodies of water contributed by man-made sources have become a matter of concern. Many bodies of water of late have exhibited biologi-cal nuisances such as dense algal and aquatic weed growths whereas in the past they supported only incidental populations of these plants. Excessive nutrients are most often blamed in the scien-tiﬁ c literature for the creation of the plant nuisances. Among the nutrients, dominant roles have been assigned by most researchers to nitrogen and phosphorus. These elements can be found in natural waters, in soils, in plants and animals, and in precipitation. Man-made sources for these nutrients are in domestic wastes and often in industrial wastes. This chapter concerns itself with the nature of algae, the environmental factors affecting their growth, the nature of the entrophication problem (sources, relative quantities of nutrients contributed by these sources, threshold limits for the growth of aquatic plants), and various techniques for the removal of those nutrients usually associated with the eutro-phication problem. THE PHYSICAL NATURE OF ALGAE Most bodies of water which can be considered eutrophic exhibit various predominant forms of algae at different times of the year. Algae that are important to investigators concerned with the eutrophication problem may be classiﬁ ed into four groups which exclude all but a few miscellaneous forms. The four groups are: 1) Blue-green algae (Myxophyceae) 2) Green algae (Chlorophyceae) 3) Diatoms (Bacillariophyceae) 4) Pigmented flagellates (Chrysophyceae, Euglenophyceae) The basis for this classiﬁ cation is the color of the organ-ism. Blue-green and green algae are self descriptive, whereas diatoms are brown or greenish-brown. Pigmented ﬂ agellates can be brown or green. They possess whip-like appendages called ﬂ agella, which permit them to move about in the water. It is not inferred by the above list that all algae are restricted to these colors. Rhodophyceae, for example, which are primarily marine algae, are brilliant red. Aquatic biologists and phytologists do not agree on the number of divisions that should be established to identify algae. Some authorities use as many as nine divisions while others use seven, ﬁ ve and four. Nevertheless, the four divi-sions as suggested by Palmer will be used as they are adequate for the ensuing discussions. BLUE-GREEN ALGAE Blue-green algae as a group are most abundant in the early fall at a temperature range of 70 to 80°F. Data obtained from water sources in the southwestern and southcentral United States indicate that for this section of the country maximum growth occurs at the end of February and through-out much of April, May and June. When blue-green algae becomes predominant, it frequently indicates that the water has been enriched with organic matter, or that previously there had been a superabundance of diatoms. Blue-green algae are quite buoyant due to the oil globules and gas bubbles which they may contain. For this and other reasons they live near the surface of the water often producing offensive mats or blankets. Since these algae are never ﬂ agellated, they are not considered swimmers although a few, such as oscillatoria and spirulina, are able to creep or crawl by body movements. Some of the common blue-green algae are anabaena, aphani-zomenon, rivularia, gomphosphaeria and desmonema. GREEN ALGAE Green algae are most abundant in mid-summer at a tem-perature range of 60 to 80°F. For water bodies in the south-western and southcentral United States, maximum growth occurs during the ﬁ rst half of September with little variation throughout the remainder of the year. Like the blue-green algae, green algae usually contain oil globules and gas bub-bles which contribute to the reasons why they are found near the surface of the water. Green algae are distinguished by their green color which comes from the presence of chloro-phyll in their cells. Many of the green algae are ﬂ agellates C005_012_r03.indd 389C005_012_r03.indd 389 11/18/2005 10:26:01 AM11/18/2005 10:26:01 AM
390
and due to their swimming ability they are frequently found in rapidly moving streams. Some of the common green algae are chlorella, spirogyra, chlosterium, hydrodictyon, nitella, staurastrum and tribonema. DIATOMS Diatoms are usually most prevalent during the cooler months but thrive over the wide temperature range of from 35 to 75°F. For water bodies in the southwestern and southcen-tral United States, diatoms thrive best in May, September and October with the maximum growth observed in mid-October. It is generally recognized that many diatoms will continue to ﬂ ourish during the winter months, often under the ice. The reason for the increase in growth twice a year is due to the spring and fall overturn, in which food in the form of carbon, nitrates, ammonia, silica and mineral matter, is brought to the surface where there is more oxygen and a greater intensity of light. Diatoms live most abundantly near the surface, but unlike the buoyant green and blue-green algae, they may be found at almost any depth and even in the bottom mud. Diatoms may grow as a brownish coating on the stems and leaves of aquatic plants, and in some cases they grow along with or in direct association with other algae. In rapidly moving streams they may coat the bottom rocks and debris with a slimy brownish matrix which is extremely slippery. Lastly, diatoms are always single-celled and nonﬂ agellated. PIGMENTED FLAGELLATES No classiﬁ cation of algae has caused more disagreement than that of the pigmented ﬂ agellates. The difﬁ culty arises from the fact that hey possess the protozoan characteristics of being able to swim by means of ﬂ agella, and the algae characteristic of utilizing green chlorophyll in association with photosynthesis. Thus they could be listed either as swimming or ﬂ agellated algae in the plant kingdom, or as pigmented or photosynthetic protozoa in the animal king-dom. One of many attempts to resolve this problem has been the proposal to lump together all one-celled algae and all protozoa under the name “Protista.” This method, however, has not met with general acceptance. For the sanitary engi-neer the motility of the organism is of lesser importance than its ability to produce oxygen. The pigmentation character-istic associated with green chlorophyll and oxygen produc-tion is sufﬁ cient criteria for separating these organisms into a class by themselves. Thus a distinction is made between pigmented ﬂ agellates (algae) and nonpigmented ﬂ agellates (protozoa). Pigmented ﬂ agellates are more abundant in the spring than at any time of the year although there is generally con-siderable variation among the individual species. Apparently ﬂ agellates are dependent on more than temperature. They are found at all depths, but usually are more prevalent below the surface of the water than at the surface. For present purposes pigmented ﬂ agellates can be divided into two groups: euglenophyceae which are grass-green in color and chrysophyceae which are golden-brown. Euglenophyceae are usually found in small pools rich in organic matter, whereas chrysophyceae are usually found in waters that are reasonably pure. Some of the more common pigmented ﬂ agellates are euglena, ceratium, mallomonas, chlamydomonas, cryptomo-nas, glenodinium, peridinium, synura and volvox. MOTILITY Of additional value in the classiﬁ cation of algae are their means of motility. Three categories have been established, namely: Nekton—algae that move by means of flagella. Plankton—algae that have no means of motility. Benthic algae—algae that attach themselves to a fixed object. NEKTON Nekton are the most active algae and are often referred to as “swimmers.” Due to their activity they use more energy and in turn release more oxygen during the daylight hours. Their cells are supplied with one, two, or more ﬂ agella which extend outward from the front, side or back of the cell. These ﬂ agella enable the organisms to move about freely in the aquatic environment and to seek food which, in the case of turbulent water, is constantly changing in location. In general nekton have the most complex structure of the three categories and come nearest to being simple animals. Nekton are the predominant algae found in swiftly moving rivers and streams. According to Lackey, results of tests performed on waters of the Ohio River show that certain nekton are the only algae that provide reliable clear-cut responses to the presence of pollution and thus are true indicator organisms. Five ﬂ ag-ellates have been singled out on the genus level as being common and easily recognized. They are (1) cryptomonas, (2) mallomonas, (3) synura, (4) uroglenopsis, and (5) dino-bryon. Dinobryon is perhaps the most easily recognized due to its unique shape which resembles a shaft of wheat. Samples taken from several rivers indicate that these algae react adversely to the presence of sewage and are found in abundance only in clean water. Unfortunately not all experts agree on what constitutes clean water and what algae serve as indicator organisms. Patrick states that the “healthy” por-tion of a stream contains primarily diatoms and green algae. Rafter states that the absence of large amounts of blue-green algae is an indicator of clean water. Palmer lists 46 species which have been selected as being representative of “clean-water algae,” and these consist of diatoms, ﬂ agellates, green algae, blue-green algae and red algae. In addition Palmer lists C005_012_r03.indd 390C005_012_r03.indd 390 11/18/2005 10:26:01 AM11/18/2005 10:26:01 AM

39147 species of algae condensed from a list of 500 prepared from reports of more than 50 workers, as being representative of “polluted-water algae.” These consist of blue-green algae, green algae, diatoms and ﬂ agellates. PLANKTON Plankton are free-ﬂ oating algae which are most commonly found in lakes and ponds, although they are by no means lim-ited to these waters. Most species are unicellular; however, they tend to become colonial when their numbers increase, as in the formation of a heavy concentrated growth known as a “bloom.” An arbitrary deﬁ nition of a bloom is that con-centration of plankton that equals or exceeds 500 individual organisms per ml. of raw water. Blooms usually show a pre-dominance of blue-green algae although algae from other classes can also form blooms. An algae bloom often becomes sufﬁ ciently dense as to be readily visible on or near the water surface, and its presence usually indicates that a rich supply of nutrients is available. Other environmental factors may stimulate the formation of blooms, and a bloom of the same organism in two bodies of water may or may not result from identi-cal favorable environmental conditions. These growths are extremely undesirable in bodies of water, in general, and in potential water supply sources in particular for the following reasons: 1) They are very unsightly. 2) They interfere with recreational pursuits. 3) When the water becomes, turbulent, fragments of the mat become detached and may enter a water treatment system clogging screens and filters. 4) When the algae die (as a result of seasonal changes or the use of algicides), decomposition occurs, resulting in foul tastes and odors. 5) They may act as a barrier to the penetration of oxygen into the water which may result in fish kills. 6) They may reduce the dissolved oxygen in the water through decay or respiration within the bloom. 7) Some blooms release toxic substances that are capable of killing fish and wild life. 8) They may cause discoloration of the water. 9) They attract waterfowl which contribute to the pollution of the water. Some of the common blue-green algae that form blooms are anabaena, aphanizomenon, oscillatoria, chlorella and hydrodictyon. Synedra and cyclotella are common diatoms that form blooms and synura, euglena and chlamydomo-nas are common ﬂ agellates that form blooms. Filamentous green plankton, such as spirogyra, cladophora and zyg-nema form a dense ﬂ oating mat or “blanket” on the surface when the density of the bloom becomes sufﬁ cient to reduce the intensity of solar light below the surface. Like blooms, these blankets are undesirable, and for the same reasons cited earlier. However, in addition, blankets also serve as a breeding place for gnats and midge ﬂ ies, and after storms they may wash up on the shores where they become offensive. In many cases hydrogen sulﬁ de and other gases which are able to spread disagreeable odors considerable distances through the air are liberated. In large amounts, hydrogen sulﬁ de has been known to seriously discolor the paint on lakeside dwellings. BENTHIC ALGAE Benthic algae are those algae which grow in close associa-tion with a supply of food. That is, they seek out an aquatic environment where nutrients are adequate, then attach themselves to a convenient stationary object such as a sub-merged twig or rock. They may be found in quiet ponds and lakes or in fast-moving rivers and streams. In some cases they break away from their attachments and form unsightly surface mats, or they may re-attach themselves somewhere else. Chlamydomonas is such an organism, where in one growth phase it may be found attached to a ﬁ xed object, and in another phase it may be dispersed throughout the water. Benthic algae include diatoms, blue-green algae, green algae and a few species of red fresh-water algae. None of the pigmented ﬂ agellates are benthic. Most attached algae grow as a cluster of branched or unbranched ﬁ laments or tubes and are fastened at one end to some object by means of an anchoring device. Others take the shape of a green felt-like mat (gomphonema), a thin green ﬁ lm or layer (phytoconis), or a soft fragile tube (tetraspora). Some of the most common benthic algae are cladophora, chara, nitella, ulothrix, cym-bella, vaucheria and gomphonema. ENVIRONMENTAL FACTORS AFFECTING GROWTH OF ALGAE The effects of certain environmental factors on the growth of the aforementioned forms of algae have been fairly well deﬁ ned. The most important parameters to be considered in the growth pattern are light intensity, temperature, pH and nutritional requirements. LIGHT INTENSITY Light is essential to all organisms which carry on photo-synthesis; however, requirements or tolerance levels differ greatly with the organism. For example, terrestrial species of vaucheria grow equally well in fully-illuminated soil and densely-shaded soil, while a number of blue-green algae grow only in shaded habitats. In addition, some algae are unable to endure in the absence of sunlight caused by sev-eral consecutive cloudy days, whereas certain submerged algae are unable to withstand exposure to full sunlight. Thus, an algae kill may be noted during a drought where C005_012_r03.indd 391C005_012_r03.indd 391 11/18/2005 10:26:01 AM11/18/2005 10:26:01 AM
392
shallow water prevents depth-dwelling algae from escaping the intensity of the sunlight penetration. Often muddy rivers are virtually algae-free due to the lack of penetration of the sunlight, the Missouri and the Mississippi being two such rivers. The distribution of algae at the various depths in a body of water is directly correlated with the intensity of illumination at the respective depths. This distribution would be difﬁ cult to express in general terms when dealing with algae on their species level. In addition, the depths at which these species would be found would change with such variables as growth phase of the organism, tempera-ture and the absorptive and reﬂ ective characteristics of the water. It can be stated, however, that certain fresh-water red algae and blue-green algae are found only at consid-erable depths and that some diatoms exist in the bottom mud. In the most general terms it can also be stated that algae are found at all levels, but most commonly near the surface. The vertical distribution may also be related to the divi-sion of light rays into various spectral colors. This division varies with the concentration of dissolved color material, plankton and particulate matter, with the seasons, and with the depth. In colored water the violet-blue end of the spec-trum is absorbed more readily. As depth increases light rays divide differently with greater absorption occurring at the red end of the spectrum. The depth to which light penetrates has a direct inﬂ uence on photosynthetic activity. The seasonal variation in this light and the resulting availability of certain dominant wave lengths may be the reason for ﬂ uctuations in the composition of the algal population from spring to fall. Much more work is needed in this area. TEMPERATURE In general, temperature is not the key factor in determining the nature of the algal ﬂ ora. Most species are able to grow and reproduce if other environmental conditions are favor-able. According to Patrick, however, the above statement is not true in the case of diatoms, where temperature changes are more important than any other environmental factor in inﬂ uencing their rate of growth. Additional work in this area by Cairns indicates that certain diatoms grow best only at a speciﬁ c temperature, and that at some temperatures they will not grow at all. Most algae are not affected by minor changes in pH brought about by the seasonal variations, growths of carbon-dioxide producing organisms, etc. Large changes such as would be caused by the introduction of industrial wastes or acid mine waters, will greatly affect algae, usually causing a decrease in population. The majority of algae thrive when the pH is near 7.0. Some blue-green algae prefer high pHs. Anacystis and coc-cochloric are found at about pH 10.0 with little or no growth below pH 8.0. Other algae such as eugleny mutabilis, cryp-tomonas erosa and ulothrix zonata prefer low pHs. NUTRITIONAL REQUIREMENTS AND TOXIC ELEMENTS FOR ALGAE Calcium Calcium is not an essential element for most algae, although some cannot develop without it. Calcium and Magnesium As bicarbonates they are a supplemental supply of carbon dioxide for photosynthesis. This accounts for the greater abundance of algae in hard-water lakes than in soft-water lakes. Iron Most algae grow best when the ferric oxide content of the water is between 0.2 to 2.0 mg per liter. Above 5 mg per liter there is a toxic effect unless it is overcome by the buffering action of organic compounds or calcium salts. Certain dia-toms (eunotia and pinnularia) are found in iron-rich water. Efﬂ uent from steel mills may be toxic to most algae if the resulting iron concentration exceeds the toxic limitation. Copper Copper is extremely toxic to algae in the range of 0.1 to 3.0 ppm as copper sulfate; the sulfate form being used as an algi-cide. Some algae are able to tolerate large amounts of copper ion and are considered copper-sulfate resistant. Protococcus, for example, is not destroyed by 10 ppm of copper sulfate. Phenol At a concentration of up to 1.9 mg per liter, phenol appar-ently has no toxic effect on diatoms. Nitrates, Phosphates and Ammonia These are essential food elements necessary for growth. Nitrogen may be obtained from nitrates, nitrites or simple ammonia compounds. The primary source of these nutrients is from sewage treatment plant efﬂ uents, although nitro-gen may be derived from the atmosphere, land runoff, etc. (See section on
.) In general as little as 0.3 to 0.015 ppm of nitrates and phosphates will produce blooms of certain species of algae, other conditions being favorable. Oil Streams polluted with oil are usually low in algae. One vari-ety of diatoms may be dominant in such waters. Salinity Increases in salinity up to about one percent do not affect the algae population. Signiﬁ cant increases, such as caused C005_012_r03.indd 392C005_012_r03.indd 392 11/18/2005 10:26:02 AM11/18/2005 10:26:02 AM

393by salt-brine wastes, may destroy most of the algae present, however. Certain fresh-water algae may become adapted to water with slowly increasing salinity. Hydrogen Sulfide At a concentration of 3.9 ppm, hydrogen sulﬁ de is toxic to most diatoms. Some resistant species are achnanthese afﬁ nis, cymbella ventricosa, hantzschia amphioxys and nitzschia palea. Silica Silica is necessary for the growth of diatoms whose cell wall is composed of silica. Presently no limits have been determined (to the author’s knowledge). Vitamins Several vitamins in small quantities are a requisite to growth in certain species of algae. Chief among these vitamins are vitamin B-12, thiamine and biotin. These vitamins are sup-plied by bottom deposits, soil runoff and by the metabolites produced by other organisms. Micronutrients Substances such as manganese, zinc, molybdenum, vana-dium, boron, chlorine, cobalt, etc. are generally present in water in the small concentrations sufﬁ cient for plant growth. Carbon Dioxide Carbon dioxide is necessary for respiration. If it is deﬁ cient, algae may remove carbon dioxide from the atmosphere. Chlorine Chlorine is toxic to most algae and is used as an algicide in the range of 0.3 to 3.0 ppm. It is used as an algicide in the treatment plant and distribution system. Some algae, cos-marium for example, are resistant to chlorine. Protococcus, which is resistant to copper sulfate, is killed by 1 ppm of chlorine. Therefore, algae resistant to the copper ion may not be resistant to the chlorine ion and vice versa. Calcium Hydroxide (lime) An excess of lime in the water, as may be introduced during pH adjustment for coagulation, results in the death of certain algae. Five ppm of lime with an exposure of 48 hours has been lethal to melosira, nitzschia and certain protozoa and crustacea. THE
PROBLEM Of the factors previously discussed which promote the growth of algae, that factor which man has altered is the nutrient concentration in may of the natural waterways. In simplest terms, eutrophication is the enrichment of waters by nutrients from natural or man-made sources. Of the many nutrients which are added to the waters by man-made sources, nitrogen and phosphorous are most often cited by researchers as being the key nutrients responsible for the promotion of algae growth. In nearly all cases when the nitrogen and phosphorus level of a body of water increases, there will be a corresponding increase in the growth of algae and aquatic plants. Such growth greatly speeds up the aging process whereby organic matter invades and gradually dis-places the water until eventually a swamp or marsh is formed. Unfortunately, the process of eutrophication is often difﬁ cult to reverse in bodies of water such as large lakes where the ﬂ ushing or replacement time for the waters can be in the order of years. The following sections provide the relative magnitude of natural and man-made sources of nutrient material associated with plant growth. SOURCES OF NUTRIENTS While it is recognized that certain algae require a number of chemical elements for growth, it is also known that algae can absorb essential as well as superﬂ uous or even toxic elements. Although every essential element must be present in algae, this does not mean that every element is essential. On the other hand, the absence of certain nutrient elements will prevent growth. Nutrients may be classiﬁ ed as (1) “absolute nutrients,” which are those which cannot be replaced by other nutrients, (2) “normal nutrients,” which are the nutrients contained in the cell during active growth, and (3) “optimum maximum growth.” It may also be well to assign a broad meaning to the word “nutrient” and deﬁ ne it as anything that can be used as a source of energy for the promotion of growth or for the repair of tissue. In evaluating the effects of nutrients on algae, care must be exercised to consider the interaction between nutrients and other physical, chemical or biological conditions. Rapid growth of algae may be stimulated more by factors of sun-light, temperature, pH, etc., than by an abundance of nutrient material. Tests performed with nitzschia chlosterium, in order to study the interaction of environmental factors showed that two identical cultures of the organism, when supplied with a reduced nutrient level, had a lower optimum light intensity and optimum temperature for maximum growth. Thus light intensity and temperature data should accompany data on nutrient concentration and growth rate. Of all the possible nutrients, only nitrogen and phosphorus have been studied in depth both in the ﬁ eld and in the labo-ratory. This is because of the relative difﬁ culties associated with the study, analysis and measurement of trace elements, compounded by the minute impurities present in the regents and distilled water. In addition, nitrates and phosphates have a long history of use in agricultural fertilizers where determination of their properties have been essential to their economical use. C005_012_r03.indd 393C005_012_r03.indd 393 11/18/2005 10:26:02 AM11/18/2005 10:26:02 AM
394

bundet

The following are the most common sources of nitrogen and phosphorus in bodies of water: 1) Rainfall—Based on experimental data, it has been found that rainwater contains between 0.16 and 1.06 ppm of nitrate nitrogen and between 0.04 and 1.70 ppm of ammonia nitrogen. Computations based on the nitrogen content of rainwater show that for Lake Mendota, Wisconsin, approximately 90,000 pounds of nitrogen are available each year as a result of rainfall. Thus it can be seen that rainfall plays a significant role in building up the nitrogen content of a lake or reservoir especially if the surface area is large. An examination of the phosphorus content of rainwater of different countries shows that a number of concentrations may exist ranging from 0.10 ppm to as little as an unmeasur-able trace, the latter reported in the Lake Superior region of the United States. In view of the wide variation in the determinations, little can be stated at present regarding the degree of phosphorus build-up in impoundments resulting from rainwater. 2) Groundwater—Studies conducted on sub-surface inflows to Green Lake, Washington, show that this water contains approximately 0.3 ppm of phospho-rus. Other reports, however, claim that the amount of phosphorus in groundwater is negligible. Investigations into the nitrate content of groundwater pro-duced variable results; however, it can be stated that 1.0 ppm is a reasonable ﬁ gure. The results of the above studies on both nitrogen and phosphorus can be summarized by stating that groundwater should not be discounted as a possible source of nutrients and that quantitative values should be obtained for the speciﬁ c locality in question. 3) Urban Runoff—Urban runoff contains storm water drainage, overflow from private disposal systems, organic and inorganic debris from paved and grassed areas, fertilizers from lawns, leaves, etc. In view of the variable concentration of the above material, precise figures cannot be obtained on the phosphorus or nitrate content that would be meaningful for all areas. Studies conducted in 1959 and 1960 by Sylvester on storm water from Seattle street gutters shows the following nutrients: Organic nitrogen—up to 9.0 ppm Nitrate nitrogen—up to 2.8 ppm Phosphorus—up to 0.78 ppm soluble and up to 1.4 ppm total. 4) Rural Runoff—Rural runoff for the purposes of definition may be considered as runoff from sparsely-populated, wooded areas with little or no land devoted to agriculture. Investigations by Sylvester showed that the phosphorus content of drainage from three such areas in the state of Washington contained 0.74, 0.77 and 0.32 lb./acre/year, or a total concentration of 0.069 ppm. The corresponding nitrate nitrogen concentration and organic nitrogen concentration amounted to 0.130 and 0.074 ppm, respectively. 5) Agricultural Runoff—Agricultural runoff is one of the largest sources of enrichment material and may be derived from two sources—wastes from farm animals and the use of nitrogen and phos-phorus-containing fertilizers. Farm-animal wastes add both large quantities and high concentrations of nutrients to adjacent streams and rivers. The large concentrations are due primarily to the practice of herding animals in relatively conﬁ ned areas. A comparison on the nutrient value of animal wastes and human wastes has been made in a study by the President’s Science Advisory Committee. According to the ﬁ ndings, a cow generates the waste equivalent of 16.4 humans, a hog produces as much as 1.9 humans and a chicken produces as much as 0.14 humans. The use of chemical fertilizers in the United States has grown almost 250% in the decade from 1953 to 1963. In 1964 the use of phosphorus-containing fertilizers and the use of nitrogen-containing fertilizers reached approximately 1.5 and 4.4 million tons, respectively, per year. Most all of this fertilizer is distributed to soil already high in natural-occurring nitrogen. When nitrogen fertilizer and natural soil nitrogen combine, there is a great increase in crop production, but also a greater opportunity for loss of this nitrogen in runoff. This loss will increase if the fertilizer is not properly applied, if it is not completely utilized by the crops, if the crops have a short growing season (the land being non-productive for a time), if the land is irrigated, and if the land is sloped. The addition of nitrogen-bearing fertilizers also increase the quantity of mineral elements in the soil runoff which are necessary for the growth of aquatic plants and algae. When applied, the nitrogen in the fertilizer is converted into nitric acid which combined with the minerals in the soil, such as calcium and potassium, rendering them soluble and subject to leaching. 6) Industrial Wastes—The nutrient content of indus-trial waste effluents is variable and depends entirely upon the nature, location and size of the industry. In some cases the effluents are totally free of nitrogen and phosphorus. The meat packing industry is one of the chief producers of nitrogen-bearing wastes. The greatest producer of phosphate-bearing wastes is most likely the phosphate-manufacturing industry itself. Most phosphate production in the United States is concentrated in Florida and as a result many severe local-ized problems of eutrophication have resulted in that state. Fuel processing industries and petroleum reﬁ neries dis-charge vast quantities of nitrogen into the atmosphere both in the gaseous state and solid state as particulate matter. This nitrogen is then washed from the atmosphere by the rain and carried back to earth. In 1964, the 500 billion tons of coal used in the United States released about 7.5 million tons of nitrogen into the atmosphere, most of which has returned to be combined with the soil. This greatly exceeds the use of nitrogen in the form of fertilizers which, as previously stated, amounted to 4.4 million tons for that year. Thus, through the atmosphere we are bringing more nitrogen into the soil than C005_012_r03.indd 394C005_012_r03.indd 394 11/18/2005 10:26:02 AM11/18/2005 10:26:02 AM

395we are taking out, and much of this excess ultimately gets washed out into our waterways. 7) Municipal Water Treatment—The water treatment plants themselves are to a degree responsible for adding to the eutrophication problem as approxi-mately 33% of the municipal water in the United States is treated with compounds containing phosphorus or nitrogen. Some of the commonly used nutrient-bearing chemicals or compounds are ammonia (in the use of chloramines) organic polyelectrolytes, inorganic coagulant aids, sodium hexametaphosphate, sodium tripolyphosphate, and sodium pyrophosphate. 8) Waterfowl—It has been estimated that wild ducks contribute 12.8 pounds of total nitrogen/acre/year and 5.6 pounds of total phosphorus/acre/year to reservoirs or lakes. A number of studies have been conducted on waterfowl, but it may be con-cluded that, although there may be some bearing on localized eutrophication, in general the overall effect is negligible. 9) Domestic Sewage Effluent—Undoubtedly the greatest contributor toward the eutrophication of rivers and lakes is the discharge from sewage treatment plants. Conventionally treated domestic sewage usually contains from 15 to 35 ppm total nitrogen and from 6 to 12 ppm total phosphorus. In addition there are a large number of minerals present in sewage which serve as micro-nutrients for algae and aquatic plants. Phosphorus in domestic sewage may be derived from human wastes, waste food (primarily from household garbage-disposal units), and synthetic detergents. Human wastes have been reported in domestic sewage at the rate of 1.4 pounds of phosphorus/capita/year. The largest source of phospho-rus, however, is from synthetic detergents which amounts to approximately 2.1 pounds/capita/year. Sawyer indicates that detergent-based phosphorus represents between 50 and 75% of the total phosphorus in domestic sewage. It should be noted that both the use of household garbage-disposal units and detergents is fairly recent, and accordingly they may be considered as contributing strongly to the development of the recently magniﬁ ed eutrophication problem. Not all the phosphorus entering a sewage treatment plant will leave the plant since chemical removal does occur during the treatment process. Calcium and metallic salts in large concentrations form insoluble phosphates which are readily removed. Very often phosphate-precipitating agents are present in waters containing industrial wastes, and when these agents are received at the plant, removals in the neigh-borhood of 60% may be realized. Nitrogen in domestic sewage is derived from human wastes and from waste food primarily from household garbage-disposal units. Human wastes, the major source of nitrogen, contributes an average of about 11 pounds of nitro-gen/capita/year. Some reduction in the nitrogen also takes place during the treatment of the sewage. Many plants treat the sludge anerobically which permits signiﬁ cant release of the nitrogen. In general the removal amounts to between 20 and 50%. The higher percentage of removal occurs when fresh wastes are given complete treatment with no return of sludge nutrients to the efﬂ uent.
STUDIES In recent years a considerable number of studies have been made on eutrophication and related factors. Most of the studies can be grouped into the following categories: 1) nutrient content of runoff, rainwater, sewage efflu-ent, bottom mud, etc. 2) nutrient analysis and physical distribution of nutrients in bodies of water before and/or after enrichment. 3) methemoglobinemia (illness in infants due to drinking high nitrate-content water) 4) toxicological and other effects on fish of high nitrate/high phosphate-content water 5) the chemical composition of plants in both eutro-phied and non-eutrophied waters 6) the nutrient values of various fertilizers, manures and other fertilizing elements 7) the nutrient value of various soils 8) the effects of eutrophication on aquatic plants, animals and fish 9) studies on specific algae under either controlled laboratory conditions or in a particular body of water, using artificial or natural environmental conditions 10) methods for the removal or reduction of nitrogen and phosphorus 11) nutrient thresholds for growth of algae and aquatic weeds 12) the effects of eutrophication on the oxygen balance. Of the above list, only studies conducted in the areas of (11) and (12) will be presented below. Work done in regard to (1) has already been presented. The removal or reduction of nitrogen and phosphorus (10) will be discussed separately as part of the subject matter in “CONTROL METHODS.” NUTRITIONAL THRESHOLDS FOR THE GROWTH OF ALGAE Studies conducted by Chu indicate that for growth on artiﬁ cial media most planktonic algae ﬂ ourish if the total nitrogen con-tent ranges from 1.0 to 7.0 ppm and the total phosphorus content ranges from 0.1 to 2.0 ppm. If the nitrogen is reduced below 0.2 ppm and the phosphorus below 0.05 ppm, the growth of algae appears to be inhibited. The same inhibiting effect is cre-ated when the nitrogen or phosphorus content is raised above 20.0 ppm. The lower limit of the optimum range of nitrogen C005_012_r03.indd 395C005_012_r03.indd 395 11/18/2005 10:26:02 AM11/18/2005 10:26:02 AM
396
varies with the organism and with the type of nitrogen. For ammonia nitrogen the optimum range varies from 0.3 to 5.3 ppm and for nitrate nitrogen the optimum range falls between 0.3 and 0.9 ppm. Below these values the growth rate decreases as the concentration of nitrogen decreases. Apparently the use of the various forms of nitrogen by algae is not constant throughout the year. Tests conducted at Sanctuary Lake in Pennsylvania (1965) indicate that the order of preference for the three forms of nitrogen—ammonia-nitrogen, nitrate-nitrogen, and nitrite-nitrogen—are deﬁ ned by three seasonal periods, which are: Spring (1) Ammonia nitrogen (2) Nitrate nitrogen (3) Nitrite nitrogen Midsummer (1) Ammonia nitrogen (2) Nitrite nitrogen (3) Nitrate nitrogen Fall (1) Ammonia nitrogen (2) Nitrate nitrogen (3) Nitrite nitrogen The amount of nitrogen in the aquatic environment is important to algae because it determines the amount of chlo-rophyll that may be formed. Too much nitrogen, however, inhibits the formation of chlorophyll and limits growth. Laboratory studies on algae conducted by Gerloff indicate that of all the nutrients required by algae, only nitrogen, phos-phorus and iron may be considered as limiting elements, and of these three, nitrogen exerts the maximum limiting inﬂ u-ence. Approximately 5 mg of nitrogen and 0.08 mg of phos-phorus were necessary for each 100 mg of algae produced. The corresponding nitrogen/phosphorus ratio is 60 to 1. Hutchinson cites phosphorus as being the more impor-tant element since it is more likely to be deﬁ cient. When phosphorus enters a body of water, only about 10% is in the soluble form readily available for algal consumption. During midsummer total phosphate may increase greatly during the formation of algal blooms, while soluble phosphate is unde-tectable due to rapid absorption by the growing algae. Very often during warm weather these blooms are stimulated by the decomposition and release of soluble phosphates from the bottom sediments, deposited by the expired blooms of previous seasons. Thus when phosphates are added to a lake, only a portion of the phosphates are used in produc-ing blooms. The blooms thrive and consume phosphates for only a short time, and a signiﬁ cant amount ﬁ nds its way to the bottom sediments where it will be unavailable to further growth of aquatic vegetation. Prescott examined a number of algae and concluded that most blue-green algae are highly proteinaceous. Aphanizomenon ﬂ os-aquae, for example, was shown to con-tain 62.8% protein. Green algae were found to be less pro-teinaceous. Spirogyra and cladophora, for example, contain 23.8 and 23.6% respectively. Thus it can be concluded that the nitrogen requirement (for the elaboration of proteins) depends on the class of algae, and that blue-green algae would require more nitrogen than green algae. Provasoli examined 154 algal species to determine the requirements for organic micronutrients. He found that although 56 species required no vitamins, 90 species were unable to live without vitamins such as B 12 , thiamin and biotin, either alone or in various combinations. He concluded that these vitamins are derived from soil runoff, bottom muds, fungi and bacterial production (B 12 ), and from a natural resid-ual in the water. Ketchum and Pirson conducted a series of examina-tions on the inorganic micronutrient requirements of algae and concluded that a number of elements are necessary for growth. No numerical values were assigned to the require-ment levels. Those elements shown to be essential were C, H, O, P, H, S, Mg, Ca, Co, Fe, K and Mo. Those elements which may be essential (subject to further study) were Cu, An, B, Si, Va, Na, Sr, and Rb. In summation, absolute values and nutrient thresholds cannot be set at this time because too little is known regard-ing the requirements of individual species. It might be stated in general terms, however, that nitrogen and phosphorus are two essential nutrient elements related to the production of blooms, and that if they are present in the neighborhood of 0.2 ppm and 0.05 ppm, respectively, algal growths will increase signiﬁ cantly. NUTRITIONAL THRESHOLDS FOR THE GROWTH OF AQUATIC PLANTS Studies conducted by Harper and Daniel indicate that sub-merged aquative plants contain 12% dry matter of which 1.8% are nitrogen compounds and 0.18% are phosphorus compounds. Hoagland indicates that when the nitrate content of water is high, nitrates may be stored in aquatic plants to be reduced to the usable ammonia nitrogen form as required. Subsequent investigations show that ammonia nitrogen can be substituted for nitrate nitrogen and used directly. Light apparently is not a necessary factor in the reduction of the nitrogen. Muller conducted a number of experiments on both algae and submerged aquatic plants, and concludes that exces-sive growths of plants and algae can be avoided in enriched waters if the concentration of nitrate nitrogen is kept below 0.3 ppm, and if the concentration of total nitrogen remains below 0.6 ppm. OXYGEN BALANCE Recently, attention has been given to the effect of the intense growths of algae on the oxygen balance of natural water-ways. It has been established that the dissolved oxygen concentrations may exhibit wide variation throughout the course of the day. This variation is attributed to the ability of algae to produce oxygen during the daylight hours, whereas they require oxygen for their metabolic processes during the hours of darkness. C005_012_r03.indd 396C005_012_r03.indd 396 11/18/2005 10:26:02 AM11/18/2005 10:26:02 AM

397 In addition, since algae are organic in nature, they exert a biochemical oxygen demand (BOD) on the stream oxygen resources as does other materials which are organic. Extensive tests were run on the Fox River in Wisconsin by Wisniewski in 1955 and 1956 to examine the inﬂ uence of algae on the puriﬁ cation capacity on rivers. In the most general terms, the studies indicate that algae increase the B.O.D. by adding organic matter capable of aerobic bacte-rial decomposition and by the respiration of the live cells which utilize oxygen during the absence of light. In the presence of light, algae produce oxygen and as a result may cause a “negative” B.O.D. for a production of oxygen in excess of that required for the normal B.O.D. require-ments or aerobic bacterial stabilization. In addition to the above, the following speciﬁ c conclusions were drawn from the tests: 1) The oxidation rate resulting from the respiration of live algae was much lower than that obtained by the biological oxidation of the dead algae. 2) The ultimate B.O.D. of live algae was practically the same as for dead algae. 3) A linear relationship was found to exist between the five-day B.O.D. of suspended matter and volatile suspended solids concentration. 4) The B.O.D. increases with increases in suspended solids, the latter consisting largely of algae. Additional work was done in this area and reported in 1965 by O’Connell and Thomas. They note that the oxygen produced by photosynthetic plants is affected greatly by changes in the availability of light due to cloud cover, turbidity in the water, etc., and therefore it may be too variable to be used as a reliable factor in evaluating the oxygen resources of a river. Another variable may be the loss of oxygen to the atmosphere during the daylight hours, caused by excess oxygen production and localized supersaturation. An important consideration is the type of photosynthetic plants which are prevalent in a river. According to the above authors, if benthic algae and/or rooted aquatic plants are predominant (in lieu of phytoplankton), there will be little beneﬁ cial effect on the oxygen balance. In addition night-time absorption of oxygen through respiration can seriously reduce daily minimum concentrations of dissolved oxygen. Determination of the effects of the benthic algae oscillato-ria along a ﬁ vemile stretch of the Truckee River in Nevada indicated that on the average of the organism produced 72.5 pounds/acre/day of oxygen through photosynthesis. Oxygen uptake for these same organisms amounted to an average of 65.4 pounds/acre/day. An examination of the oxygen proﬁ les indicated that the oxygen variation throughout the day ranged from 2 (at night) to 13 (during daylight) parts per million. It is dissolved oxygen variations such as the above which has been responsible for the disappearance of high quality game ﬁ sh in many of our natural waterways. CONTROL METHODS TO PREVENT
There are a number of methods which attempt to limit the amounts of nutrients in bodies of water once the point of eutrophy has been reached. Some of these include dredging and removing bottom sediments with an inert liner, harvesting the algae, ﬁ sh, aquatic weeds, etc., and diluting the standing water with a water of lower nutrient concentration. Although these methods may have their proper application, if eutrophi-cation is to be decelerated, nutrient removal must start before wastes are permitted to enter the receiving waters. Regarding the speciﬁ c nutrients necessary to be removed, most researchers have placed the blame of eutrophication in waters to the inorganic forms of phosphorus and nitrogen. A smaller number of researchers are claiming that the algae–bacteria symbiosis relationship might be responsible for the rapid growth of blooms and that the amount of algae pres-ent in natural waters is in direct balance with the amount of carbon dioxide and/or bicarbonate ions in the waters. They further argue that an external supply of the above elements is necessary for the growth of algae populations. Since nei-ther theory has been proved conclusively to date, the control methods given will be for the removal of nitrogen and phos-phorus since it is these nutrients which most researchers lay to the blame of eutrophication and which have been there-fore subsequently studied in detail. NITROGEN REMOVAL Land Application It has been found that nitrogen-bearing waters, when perco-lated through soil are subjected to physical adsorption and biological action which removes the nitrogen in the ammo-nium form. It appears, however, that the nitrate form of nitrogen remains unaffected. At present this process is only at the theoretical stage, and to the author’s knowledge no full-scale application has been attempted. Considerable land area would be involved which may prove a deterrent. Anaerobic Denitrification In this process, the nitrate present in sewage is reduced by denitrifying bacteria to nitrogen and nitrous oxide gases which are allowed to escape into the atmosphere. In order to satisfy the growth and energy requirements of the bacteria, methanol in excess of 25 to 35% must be added as a source of carbon. The removal efﬁ ciency ranges from 60 to 95%. The major advantage to anaerobic denitriﬁ cation is that there are no waste products requiring disposal. This process is still primarily in the experimental stage at this date. Ammonia Stripping Ammonia stripping is an aeration process modiﬁ ed by ﬁ rst raising the pH of the wastewater above 10.0. At this pH the C005_012_r03.indd 397C005_012_r03.indd 397 11/18/2005 10:26:02 AM11/18/2005 10:26:02 AM
398
ammonia nitrogen present is readily liberated as a gas and is absorbed into the atmosphere. Aeration is usually accom-plished in a packed tray tower through which air is blown. This process is suited to raw sewage where most of the nitrogen is either in the ammonia form or may be readily converted to that form. In secondary treatment processes the conversion of ammonia nitrogen to nitrate nitrogen can be retarded by maintaining a high organic loading rate on the secondary process. Efﬁ ciency of nitrogen removal by ammonia stripping is excellent with 80 to 98% reported. There is also the advantage that there are no waste materials which must be disposed of. PHOSPHORUS REMOVAL Chemical Precipitation Precipitation of phosphorus in wastewater may be accom-plished by the addition of such coagulants as lime, alum, ferric salts and polyelectrolytes either in the primary or sec-ondary state of treatment, or as a separate operation in ter-tiary treatment. In general, large doses in the order of 200 to 400 ppm of coagulant are required. However, subsequent coagulation and sedimentation may reduce total phosphates to as low as 0.5 ppm, as in the case of lime. Doses of alum of about 100 to 200 ppm have reportedly reduced orthophos-phates to less than 1.0 ppm. Phosphorus removal by chemical coagulation generally is efﬁ cient with removals in the order of 90 to 95% reported. Additional beneﬁ ts are gained in the process by a reduction in B.O.D. to a value of less than 1.0 ppm. Both installation and chemical costs are high, however, and the sludges pro-duced are both voluminous and difﬁ cult to dewater. Sorption Sorption is the process of passing wastewater down-ward through a column of activated alumina whereby the common form of phosphate are removed by ionic attraction. Regeneration of the media is accomplished by backwash-ing with sodium hydroxide followed by acidiﬁ cation with nitric acid. Contrary to alum treatment, this process has the advantage in that sulfate ions are removed and thus the sulfate concentra-tion is not increased. Since no salts are added, the pH and the calcium ion concentration remain unchanged. The process is efﬁ cient with more than 99% removal reported. The process should be limited to wastewater with a moderate amount of solids so as not to clog the media. REMOVAL OF NITROGEN AND PHOSPHORUS Biological (secondary) Treatment In the secondary method of sewage treatment, bacteria uti-lize soluble organic materials and transform them into more stable and products. In the process nitrogen and phosphorus are removed from the wastes, utilized to build new cellular materials, and the excess is stored within the cell for future use. For each pound of new cellular material produced, assuming the material to be in the form of C 5 H 7 NO 2 , about 0.13 pounds of nitrogen and about 0.026 pounds of phospho-rus would be removed from the sewage. In the actual opera-tion of this process not all of this nitrogen is removed unless additional energy material in the form of carbohydrates is added. Although it may be possible to eliminate all the nitrogen, a considerable amount of soluble phosphorus may remain, possibly because of the high ratio of phosphorus to nitrogen in sewage, attributable to synthetic detergents. Much of this phosphorus can be removed by absorption on activated sludge ﬂ oc when it is later separated and removed. This process offers a 30 to 50% removal of nitrogen and about a 20 to 40% removal of phosphorus without the spe-cial addition of carbohydrates. Reverse Osmosis The process of reverse osmosis consists of passing wastewa-ter, under pressures as high as 750 psi, through a cellulose acetate membrane. The result is the separation of water and all ions dissolved therein. In actual practice the process has been plagued with difﬁ culties primarily due to membrane fouling or premature failure of the membrane. In addition some nitrate and phosphate ions escape through the membrane. Removal efﬁ ciency ranges from between 65 to 95% (for nitrogen). Electrodialysis Like reverse osmosis, electrodialysis is a non-selective demineralization process which removes all ions which would include the nitrate and phosphate ions. Essentially an electric current is used in conjunction with a membrane inserted in the line of current ﬂ ow to separate the cations and anions. The problems that have developed in the operation of this process include membrane clogging and precipitation of low-solubility salts of the membrane. Acidiﬁ cation of the water and removal of some of the solids prior to treatment has been effective in minimizing these problems, although it adds to the cost. Removal efﬁ ciency ranges from between 30 to 50% (for nitrogen). Ion Exchange In the ion exchange process wastewater is passed through a media bed which removes both anionic phosphorus and anionic nitrogen ions and replaces them with another ion from the media. Regeneration of ion exchangers is com-monly accomplished with inexpensive sodium chloride, and frequently the salt is salvaged by recycling the backwash water. Difﬁ culties in the process may be caused by fouling of the exchange resin due to organic material and reduction in C005_012_r03.indd 398C005_012_r03.indd 398 11/18/2005 10:26:02 AM11/18/2005 10:26:02 AM

399the exchange capacity due to sulfates and other ions. The former may be reduced by removing the organic matter from the resin with sodium hydroxide, hydrochloric acid, methanol and bentonite. The efﬁ ciency and cost of nitrogen and phosphorus removal by ion exchange depends largely on the degree of pretreatment and/or the quality of the water to be treated. Removal of nitrogen ranges from between 80 to 92%. A number of ions exchange resins are available for nitro-gen removal alone. These include zeolites, strong base anion resins (Amberlite IRA-410) and nuclear sulfonic cation resins (Nalcite HCR and Amberlite IR-120). Algae Harvesting A two-phase process which involves (1) growing algae in special shallow wastewater ponds where they feed on the absorb nutrients, and (2) removing the algae which then con-tain the nutrients within their systems. Algal predominance will depend upon the type of nutrients available and the con-centrations. Frequently the ﬂ agellates euglena and chlorella will predominate where the nutrient concentration is high, and ﬁ lamentous green algae such as spirogyra, vaucheria and ulothrix will predominate where the nutrient concentra-tion is more moderate. The most desirable algae would be those that are large, those that would grow rapidly and those that would require vast quantities of food for energy, such as the swimming algae. In addition to removing nutrients, algae produce oxygen which reduces the B.O.D., and certain ﬂ agellates which ingest inorganic solids, are able to stabilize some of the organic material. One of the major difﬁ culties experienced with this process involves the harvesting procedure. A number of methods have been tried which include screening, settling, centrifuging and chemical screening. All have been found to present some form of difﬁ culty, although it appears from the standpoint of performance and economy, the screening method may be the least unsatisfactory. Another problem is that complete nitrogen removal is seldom achieved unless a carbon source, such as carbon dioxide or methanol, is sup-plied. Still another problem is the need for disposing of a huge, sloppy mass of slimy and odoriferous dead algae. Pilot studies show that a high-rate continuous-ﬂ ow pro-cess is feasible when light is not limiting, and that orthophos-phate concentrations can be reduced 90% to less than 1.0 ppm within 6 to 12 hours. Nitrogen removal is variable with estimates ranging from 40 to 90% efﬁ ciency depending upon the feed rate, pond design, and climatic conditions. The major drawbacks to this process aside from those mentioned, are the large land requirements needed and the necessity to rely on climatic conditions. In the latter case, artiﬁ cial illumination may prove of value depending on power rates in the area. CONCLUSIONS AND OBSERVATIONS It might be concluded after examining a vast array of mate-rial that the bio-physical and bio-chemical factors which affect algae are extremely complex, and it is difﬁ cult to pre-dict, with exacting certainty, future events relating to algal growth. The complexity is greatly magniﬁ ed by the interac-tion which apparently exists between the numerous factors themselves. When these factors are examined and evaluated, the conclusions reached by one observer are not always in complete agreement, or may disagree entirely, with the con-clusions reached by another observer. Commonly, speciﬁ c organisms may produce different reactions because of their phase of life, seasonal changes, or because of other complex and little understood metabolic functions. In few areas is there less accord than in the literature written about algae. (See references—The Physical Nature of Algae, #1 to 33, at the end of this article.) Authorities vary sharply in their opinions as to classiﬁ cation, physical descriptions, toxicity thresholds, etc. When this disagree-ment is added to the previously expressed uncertainties, it can be seen that our current knowledge is subject to various interpretations. Thus it is evident that there is a need for more work; work to develop new and useful information which will receive universal acceptance and which will clarify and expand our present knowledge. Conferences dealing with unique and local problems involving algae and eutrophica-tion are ongoing. (See References—Eutrophication, #53, 55, 56 and 57, at the end of this article.) ACKNOWLEDGMENTS My sincere appreciation is extended to Mr. Robert G. Wieland who helped in the preparation of the manuscript and to the Research Foundation of the Newark College of Engineering for their aid in the typing of the manuscript. REFERENCES THE PHYSICAL NATURE OF ALGAE 1. Palmer, Mervin C., Algae in Water Supplies, Robert, A., Taft Sanitary Engineering Center, Cincinnati, Ohio, Public Health Service Pub. No. 657, 1959, p. 8. 2. Babbitt, H.E. and J.J. Doland, Water Supply Engineering, McGraw-Hill, N.Y., 1955, p. 457. 3. Kudo, R. R., Protozoology, Chas. C. Thomas Publisher, 3rd Ed., 1950. 4. Palmer, Mervin C. and W.M. Ingram, Suggested classification of algae and protozoa in sanitary science, Sewage and Industrial Wastes, 27, 1955, pp. 1183–1188. 5. Silvey, J.K.G. and A.W. Roach, Studies on microbiotic cycles in sur-face waters, Jour. of the Amer. Water Works Assn., Jan., 1964, p. 61. 6. Lackey, James B., Two groups of flagellated algae serving as indicators of clean water, Jour, of the Amer. Water Works Assn., June, 1941, p. 1000. 7. Patrick, Ruth. A proposed biological measure of stream conditions, Proc. of the 5th Industrial Waste Conference, Purdue, Univ. Engineer-ing Bull. No. 34, 1950, pp. 379–399. 8. Rafter, G.W., The Microscopial Examination of Potable Water, Van Nostrand Co., N.Y., 1960. 9. See Reference 1. above, pp. 41 and 42. 10. See Reference 1. above, p. 38. 11. Lackey, J.B., Plankton as related to nuisance conditions in surface waters, Amer. Assn, for the Advancement of Science, 1949, pp, 56–63. 12. Krauss, Robert W., Fundamental characteristics of algal physiology, Sem-inar on Algae and Metropolitan Wastes, Robert, A., Taft Inst., Cincinnati, Ohio, Apr., 27–29, 1960, p. 41. C005_012_r03.indd 399C005_012_r03.indd 399 11/18/2005 10:26:02 AM11/18/2005 10:26:02 AM
400

Ingram, W.M. and G.W. Prescott, Toxic fresh-water algae, Amer. Midland Naturalist, 52, 1954, pp. 75–87. 14. Derby, R.L. and D.W. Graham, Control of aquatic growths in reservoirs by copper sulfate and secondary effects of such treatment, Proc. Amer. Soc. of Civil Engineers, Vol. 79, 1953. 15. Borchardt, J.A., Eutrophication causes and effects, Jour. of the Amer. water Works Assn., June 1969, pp. 272–275. 16. Thomas, S. and J.W. Burger, Chlamydomonas in storage reservoirs, Jour. of the Amer. Water Works Assn., July, 1933. 17. Smith, Gilbert M., The Fresh-Water Algae of the United States, McGraw-Hill, N.Y., 1933, p. 13. 18. Mackenthum, K.M. and W.M. Ingram, Limnological Aspects of Recreational Lakes, U.S. Public Health Service Pub. No. 1167, 1964, pp. 18–19. 19. Lauff, George H., The role of limnological factors in the availability of algae nutrients, Contrib. No. 22, Univ. of Georgia Marine Inst., Sapelo Is., Ga. 20. Patrick, Ruth, Factors effecting the distribution of diatoms, Botanical Review No. 14, 1948, pp. 473–524. 21. Cairns, J., Jr., Effects of increased temperatures on aquatic organisms, Industrial Wastes, 1, 1956, pp. 150–152. 22. Gerloff, G.C. et al., The mineral nutrition of coccochlores peniocystis, Amer. Jour. of Botany, 37, 1950, pp. 835–840. 23. Lackey, J.B., Aquatic life in waters polluted by acid mine waste, U.S. Public Health Service Report No. 54, 1939, pp. 740–746. 24. Claassen, P.W., Biological studies of polluted areas in the Genesee river system, N.Y. State Dept. of Conservation, Supplement to the 16th Annual Report for 1926, 1927. 25. Water Quality and Treatment, Amer. Water Works Assn., N.Y., 1951, p. 119. 26. See Reference 1, above, p. 57. 27. Lackey, J.B., Algal density as related to nutritional thresholds, Seminar on Algae and Metropolitan Wastes, Robert, A., Taft Inst., Cincinnati Ohio, Apr., 27–29, 1960, p. 59. 28. See Reference 17, above, p. 15 (see reference to Moore, 1917, and Moore and Carter, 1923). 29. See Reference 17, above, p. 57. 30. Provasoli, Luigi, Micronutrients and heterotrophy as possible factors in bloom production in natural waters, Seminar on Algae and Metropolitan Wastes, Robert, A., Taft Inst., Cincinnati, Ohio, Apr., 27–29, 1960, p. 48. 31. See Reference 25, above, pp. 112–113. 32. See Reference 1, above, p. 65. 33. See Reference 25, above, p. 111.
1. Fruh, Gus E., Biological responses to nutrients—eutrophication: Prob-lems in fresh water, Water Resources Symposium No. 1, Univ. of Texas, Austin, 1968. 2. Ketchum, B.H., Mineral nutrition of phytoplankton, Annual Review of Plant Physiology, No. 5, 1954, p. 55. 3. Maddux, W.S. and R.F. Jones, Some interactions of temperature, light, intensity and nutrient concentration during the continuous culture of Nitzschia Closterium and Tetraselmis sp., Jour. of Limnology and Oceanography, Vol. 9, 1964. 4. Carrol, D., Rainwater as a chemical agent of geological progress—A review, Geological Survey Water—Supply 1535-G, 1962. 5. Fruh, Gus E. and G.F. Lee, The effect of eutrophication upon the water resources management of the Yahara river basin, report of the Water Chemistry Program, Univ. of Wisconsin, Madison, Wisc. 6. Putnam, H.D. and T.A. Olson, an investigation of nutrients in western lake Superior, School of Public Health Report, Univ. of Minnesota, 1960. 7. Sylvester, R.O. and G.C. Anderson, A lake’s response to its environ-ment, American Society of Civil Engineers, Sanitary Engineering Div., 90, 1964. 8. Sawyer, C.N., Causes, effects and control of aquatic growths, Jour. of the Water Pollution Control Fed., 34, 1962, 279. 9. Reference 1, above, p, 51. 10. Sylvester, R.O., An engineering and ecological study for the rehabilita-tion of Green Lake, Univ. of Washington, Seattle, Wash., 1960. 11. Restoring the quality of the environment, report of the Environmental Pollution Panel, President’s Science Advisory Committee, The White House, Wash., D.C., 1965. 12. Sauchelli, V., Chemistry and Technology of Fertilizers, ACS Mono-graph Series No. 148, Reinhold Pub. Co., N.Y., 1960. 13. Sources of nitrogen and phosphorus in water supplies, Task Group Report, Jour. of the Amer. Water Works Assn., Mar. 1967, p. 347. 14. Andrews, W.R., The Response of Crops and Soils to Fertilizers and Manures, State College, Mississippi, 1947. 15. Reference 13, above, p. 349. 16. Reference 13, above, p. 350. 17. Paloumpis, A.A. and W.C. Starrett, An ecological study of benthic organ-isms in three Illinois river flood plain lakes, Amer. Midland Naturalist, Vol. 64, 1906. 18. Hume, N.B. and C.E. Gunnerson, characteristics and effects of hyper-ion effluent, Jour. of the Water Pollution Control Fed., 34, 1962, p. 15. 19. Hawk, P.B. et al., Practical Physiological Chemistry, 13th Ed., The Blakiston Co., Phila., Pa., 1955. 20. Sawyer, C.N., Problems of phosphorus in water supplies, Jour. of the Amer. Water Works Assn., Nov., 1965, p. 1431. 21. Hurwitz, E. et al., Phosphates—Their fate in a sewage treatment plant waterway system, Water and Sewage Works, 112, 1965, p. 84. 22. Standard practice in separate sludge digestion—committee report, Proc. of the Amer. Soc. of Civil Engineers, 63, 1937, p. 39. 23. Reference 13, above, p. 348. 24. Mackenthun, Kenneth M., Nitrogen and Phosphorus in Water, An Annotated Selected Bibliography of Their Biological Effects, U.S. Public Health Service, 1965. 25. Chu, S.P., The influence of the mineral composition of the medium on the growth of planktonic algae—Part I, Jour. of Ecology, 30, No. 2, 1942, pp. 284–325. 26. Chu, S.P., The influence of the mineral composition of the medium on the growth of planktonic algae—Part II, Jour. of Ecology, 31, No. 2, 1943, pp. 109–148. 27. Dugdale, V.A. and R.C. Dugdale, Nitrogen metabolism in lakes III; Tracer studies of the assimilation of inorganic nitrogen sources, Limnology and Oceanography, 10, No. 1, pp. 53–57. 28. Gerloff, G.C. and F. Skoog, Nitrogen as a limiting factor for the growth of microcystis aereginosa in southern Wisconsin Lakes, Jour. of Ecology, 38, No. 4, 1957, pp. 556–561. 29. Hutchinson, G.I., A Treatise on Limnology, John Wiley, N.Y., 1957. 30. Prescott, G.W., Biological disturbances resulting from algal populations in standing waters, the ecology of algae, Spec. Pub. No. 2, Pymatuning Laboratory of Field Biology, Univ. of Pittsburgh, 1959, pp. 22–37. 31. Provasoli, Luigi, Micronutrients and heterotrophy as possible factors in bloom production in natural waters, algae and metropolitan wastes, Trans. of the 1960 Seminar, Taft Sanitary Engineering Center, Cincin-nati, Ohio. 32. Reference 2, above, pp. 55–74. 33. Harper, H.J. and H.R. Danial, Chemical composition of certain aquatic plants, Botanical Gazette, 96, 1939, p. 186. 34. Hoagland, D.R., Lectures on the inorganic nutrition of plants, Chronica Botanica Co., Waltham, Mass., 1944. 35. Muller, W., Nitrogen content and pollution of streams, Gesundheits-ing., Water Pollution Abstracts, 28, No. 2, Abs. No. 454, 74, p. 256. 36. Wisniewski, Theodore F., Algae and their effects on D.O. and B.O.D., Parts I–III, water and sewage works, June–Aug., 1958. 37. O’Connell, R.L. and N.A. Thomas, Effects of benthic algae on stream dissolved oxygen, Jour. of the Sanitary Engineering Div. — Amer. Soci-ety of Civil Engineers, June, 1965, pp. 1–16. 38. Lange, Willy, Effect of carbohydrates on the symbiotic growth of planktonic blue-green algae with bacteria, Nature, Sept., 1967. 39. Kuentael, L.E., Bacteria, carbon dioxide and algal blooms, Jour. of the Water Pollution Control Fed., October, 1969. 40. King, D.L., The role of carbon in eutrophication, Jour. of the Water Pollution Control Federation, Dec., 1970. 41. Herbert, C.P. and G.S. Schroepfer, The travel of nitrogen in soils, Jour. of the Water Pollution Control Fed., 39, No. 1, Jan., 1967. C005_012_r03.indd 400C005_012_r03.indd 400 11/18/2005 10:26:03 AM11/18/2005 10:26:03 AM

bundet

401 42. Eliassen, Rolf et al., Removal of nitrogen and phosphorus, 23rd Purdue Industrial Waste Conference, Purdue Univ., Lafayette, Ind., May, 8, 1968. 43. Missingham, G.A., Occurrence of phosphorus in surface waters and some related problems, Jour. of the Amer. Water Works Assn., Feb., 1967, p. 207. 44. Culp, R.L., Wastewater reclamation by tertiary treatment, Jour. of the Water Pollution Control Fed., 35, No. 6, June, 1963. 45. Reference 20. above, p. 1438. 46. Yee, W.C., Selective removal of mixed phosphates by activated alu-mina, Jour. of the Amer. Water Works Assn., Feb., 1966, p. 246. 47. Bennett, G.E. et al., Water reclamation study program, Water Quality Control Lab., Stanford Univ., Stanford, Calif., Aug., 1968. 48. Rolich, G.A., Chemical methods for the removal of nitrogen and phos-phorus from sewage treatment plant effluents, algae and metropolitan wastes, Trans. of the 1960 Seminar, Taft Sanitary Engineering Center, Cincinnati, Ohio. 49. McKinney, Ross E., Microbiology for Sanitary Engineers, McGraw-Hill, N.Y., 1962, p. 243. 50. Bogan, R.H., The use of algae in removing nutrients from domestic sewage, algae and metropolitan wastes, Trans. of the 1960 Seminar, Taft Sanitary Engineering Center, Cincinnati, Ohio. 51. Jones, R.A. and G.F. Lee, 1982. Recent advances in assessing the impact of phosphorus loads on eutrophication-related water quality. Water Res., 16: 503–515. 52. Rast, W., R.A. Jones and G.F. Lee. 1983. Predictive capability of US OECD phosphorus loading/eutrophication response models. Journ. Water Pollut. Control Fed. 55, 990–1003. 53. White, E., 1983. Lake eutrophication in New Zealand—A comparison with other countries of the Organization for Economic Cooperation and Development, New Zealand Journal of Marine and Freshwater Research, 17, 437–444. 54. Birge, W.J., J.A. Black and A.G. Westerman, “Short-Term Fish and Amphibian Embryo-Larval Tests for Determining the Effects of Toxi-cant Stress on Early Life Stages and Estimating Chronic Values for Single Compounds and Complex Effluents,” Environmental Toxicology and Chemistry, Vol. 4, 1985, pp. 807–821. 55. Brown, O.B., R.H. Evans, J.W. Brown, H.R. Gordon, R.C. Smith and K.S. Baker, “Phytoplankton Blooming Off the US East Coast: A Satel-lite Description,” Science, 229: 163–167 (1985). 56. Suszkowski, D. J. and J.M. Mansky, “The Disposal of Sediments Dredged from New York Harbors,” Management of Bottom Sediments Containing Toxic Substances, 10th U.S./Japan Experts Meeting, 1985. 57. New Jersey Department of Environmental Protection (NJ DEP). “Green Tide: Occurrence, Identification and Impacts,” report of the New Jersey Department of Environmental Protection (1987). ROBERT DRESNACK New Jersey Institute of Technology C005_012_r03.indd 401C005_012_r03.indd 401 11/18/2005 10:26:03 AM11/18/2005 10:26:03 AM
402F
INTRODUCTION The technology for reacting suspended coal particles with a gas ﬂ owing through them dates back to the 1920s when the Winkler gas generator was developed in Germany. The petroleum industry was responsible for the commercial expansion of ﬂ uidization techniques in the U.S. (1940s), particularly in the use of solids which catalytically crack vaporized heavy oils to produce gasoline and other petro-leum fuels. The application of ﬂ uidized bed combustion (FBC) technology (to various solid fuels) is widespread in the U.S. and in other countries for all types of industrial processes. More than 350 atmospheric ﬂ uidized bed units are operating in North America, Europe and Asia. FBC is part of the answer to the question—how do we control our major emissions from coal sources? Brieﬂ y an FBC boiler is a ﬁ nely divided bed of solid fuel particles in admixture with limestone particles which are suspended or conveyed by primary combustion air moving in the vertical upward direc-tion. The limestone reacts with sulfur dioxide to remove it from the ﬂ ue gas. The low uniform temperature (ca 1550⬚F) has a beneﬁ cial effect on nitrogen oxide suppression. The emission from coal combustion schemes of nitrogen oxides (NO x ) and sulfur dioxide (SO 2 ), together with carbon oxides (CO and CO 2 ), particulate matter and solid wastes must always be compared when evaluating various alternative schemes. The potential consequences of gaseous emissions, include the greenhouse effect and acid rain, which have received much publicity in recent years. The practical FBC limit of SO 2 removal is currently about 95%. Nitrogen oxide formation is lower than with conventional pulverized coal (PC) boiler NO x control. TYPES OF FLUIDIZED BED COMBUSTORS (FBCS) FBCs are generally referred to as either circulating (CFB) or bubbling beds. However, the bubbling type may be classiﬁ ed according to whether reaction takes place at atmospheric (AFB) or under pressurized conditions (PFB). A. Circulating Fluidized Bed Combustors In the basic CFB combustor, coal or some other type of fossil fuel, e.g., natural gas or petroleum, is injected into the com-bustor together with a calcium based material such as lime-stone or dolomite to be used as a sorbent for SO 2 . The bed material is entrained by ﬂ uidizing air usually in the velocity range of 12–30 ft/sec. The entrained material is forced into a refractory-lined cyclone located between the combustor and the convective pass. The separated larger particles are reintroduced at the bottom of the combustion chamber or, as in some designs, to an external heat exchanger. The mean bed particle size is usually between 50 and 300 microns. Combustion temperature will vary but generally is kept between 1550⬚F and 1650⬚F. 1 In this temperature range SO 2 sorption is optimized and the formation of nitrogen oxides is minimized. The heavier solids fall to the bottom of the cyclone and are recirculated at a ratio of between 15:1 and 100:1 (solids to feed). The carbon content of the bed is usually about 3–4%. Calcium sulfate, ash, and calcined limestone make up the bulk of the recirculated material. The ﬂ ue gas exits the top of the cyclone, travels through the convective pass and typically goes into an economizer (heat exchanger—superheated steam produced) and into a tubular air preheater. From there the gas may enter an electrostatic precipitator or a bag house dust collector (for removal of ﬁ ne particulate matter from the gas). An induced draft fan is ﬁ nally employed to force the gas up a stack and into the atmosphere. Combustion air is provided at two levels of the combus-tor. Primary air enters through the bottom of the combustor and is evenly distributed by a gas distributor plate. Secondary air enters through a number of ports in the sidewalls of the combustor. Hence, there are two staged areas of combustion within the combustor. In the lower combustor, combustion takes place under reducing conditions. In the upper com-bustor nitrogen oxides are further reduced as is particulate matter. The admission of secondary air is also beneﬁ cial in controlling the temperature of the combustor as well as in C006_001_r03.indd 402C006_001_r03.indd 402 11/18/2005 2:28:14 PM11/18/2005 2:28:14 PM

403maintaining the transport (entrainment) of the bed material throughout the length of the combustor. The density of the bed naturally varies with the combustor height, with density increasing towards the bottom. Steam may be produced at several locations. Water-walls ﬁ xed to the upper portion of the combustor extract heat gen-erated by the combustor. The convective pass also emits heat associated with the hot ﬂ ue gas and solids which pass through it. External heat exchangers are also employed for the steam production. These heat exchangers (EHE’s) are unﬁ red, dense ﬂ uidized beds, which extract heat from the solids which fall to the bottom of the cyclone(s). More than one cyclone may be employed. The heat exchange is accom-plished before the material is returned to the combustor. The external heat exchanger is a device which can, thus, be used as an effective additional method for controlling combustor temperature. The heat transfer coefﬁ cients to the water-walls usually lie between 20 and 50 Btu/hr. ft. 2 ⬚F. B. Bubbling Fluidized Bed Combustors Bubbling ﬂ uidized bed combustors are characterized by distinct dense beds. The bed material may be recirculated as in the case of CFB’s, but at substantially lower recycle ratios (between 2:1 and 10:1). Particle velocities are usu-ally between 2–15 ft/sec and a small amount of bed material is separated out (elutriated) as compared with CFB’s. The mean bed particle size generally lies between 1000 and 1200 microns. As with CFB’s, the fuel used is usually coal or some other type of fossil fuel. Limestone or some other sorbent material is also used to decrease SO 2 emissions. The feed material may be fed either over the bed or under the bed. The manner of the feeding is an important design criterion in that it effects boiler control, emissions control (especially for SO 2 ) and combus-tion efﬁ ciency. Many bubbling bed designs incorporate over-bed feeding in which the feed is “thrown” into the combustor by pressurized air. This overbed method can often be a disad-vantage because throwing distance is limited. Hence, a long, narrow boiler is often required. The underbed method of feeding is often associated with plugging and erosion problems. However, these problems can be avoided with proper design considerations. The Tennessee Valley Authority (TVA) has designed a 160 MW bubbling bed unit at its Shawnee station in Kentucky. The facility was constructed at a cost of $232,000,000 (1989). EPRI believes that most retroﬁ ts would fall into the $500–1000/kW range (1989 dollars) and that the levelized generation cost would be 5–10% less than a conventional unit with downstream ﬂ ue gas treatment. 1a The coal used for this unit is crushed to less than ¼ inch and dried with ﬂ ue gas to less than 6% moisture. The fuel then passes through a ﬂ uidized bottle splitter with a central inlet and fuel lines arranged concentrically around the inlet. The feed material is forced into the combustor from the bottles, which are pressurized, by blowers. Each bottle acts as an individual burner and can be used to control load in the same way as cutting a burner in and out. 2 When overbed feeding is used, the ﬁ ne material in the fuel has a tendency to elutriate too swiftly. If the fuel is fed underbed, the ﬁ nes will have a longer residence time. Excess CO generation can result with the excessive burning of ﬁ nes. This in turn can lead to overheating which could cause superheater controls to trip-off. Ash-slagging is another potential problem asso-ciated with overheating. Sometimes it may be necessary to recycle the ﬂ y-ash in order that carbon is more thoroughly burned and sorbent more completely utilized. In-bed combustor tubes are generally used to extract heat (create steam). The heat transfer coefﬁ cient range is higher than that of CFB’s, i.e., 40–70 Btu/hr.ft 2 ⬚F. Erosion of the tubes is a problem which is ever present in the bubbling bed combustor. The problem worsens as bed particle velocity increases. Horizontally arranged tubes are more susceptible to erosion than are vertical tubes. Various methods of erosion protection include metal spray coatings, studding of the tube surfaces with small metal balls, and wear ﬁ ns. Occasionally recycled cold ﬂ ue gas is used in lieu of tubes. Waterwalls located in the upper portion of the combustor are also used (as with CFB’s) to extract heat. The lower portion is refractory lined. Combustor free-board is usually between 15 and 30 ft. The typical convective pass, cyclone, air heater, particle separator scenario closely resembles that of the CFB. C. Pressurized Fluidized Bed Combustors (PFBCs) The pressurized ﬂ uidized bed combustor is essentially analo-gous to the bubbling bed combustor with one exception—the process is pressurized (10 to 16 atmospheres) thereby allow-ing the ﬂ ue gas to drive a gas turbine/electric generator. This gas turbine along with a stream-driven turbine creates a very efﬁ cient “combined cycle” arrangement. PFBCs may also be “turbocharged,” i.e., before the ﬂ ue gas enters the gas turbine, heat is extracted via a heat exchanger. Steam created by the energy transfer is used to drive the compressor which pressurizes the system. There is no energy excess to drive an electric generator in this case. Deeper beds (typically 4 m.) may be used in PFBCs because they are pressurized. The residence time of a parti-cle in the bed is longer than that of a particle in the shallower bed of a bubbling bed combustor. The ﬂ uidizing velocity (typically 1 m/s) is also lowered because of pressurization. As mentioned before, lower velocities minimize the amount of in-bed tube erosion. Two other beneﬁ ts of pressurization are a reduced bed cross-sectional area and reduced boiler height. Since combustor efﬁ ciency and sorbent utilization are excellent, recycle is rarely needed. However, when very unre-active fuels are burned, recycling of ﬁ nes may be necessary. Since PFBCs are pressurized, certain design character-istics must be taken into consideration, especially in regard to the gas turbine. This turbine supplies the combustion and ﬂ uidizing air for the bed. Unlike conventional AFBC’s the turbine inlet air is dependent upon certain temperature and pressure conditions since this inlet air is actually the exhaust gas from the combustor. To compensate for variations in load and subsequent changes in the exhaust gas conditions the gas C006_001_r03.indd 403C006_001_r03.indd 403 11/18/2005 2:28:15 PM11/18/2005 2:28:15 PM
404
turbine must be ﬂ exible. An effective turbine should be able to accept low gas temperatures, be minimally affected by unremoved ﬁ nes in the gas, compensate for low load condi-tions, and allow the gas velocity through the hot gas clean up (HGCU) system and excess air to remain near constant over much of the load range. Most FBC systems incorpo-rate a free-wheeling low pressure and constant velocity high pressure shaft design to accomplish the aforementioned requirements. The HGCU system generally consists of one or several cyclones. Sometimes a back-end ﬁ lter at conven-tional pressuers and temperatures is used in addition. The gas turbine accounts for approximately 20% of a FBC’s total power output while the steam turbine creates the remainder. 5 The steam turbine is powered from steam created via combustor tubes and is totally independent of the exhaust gas and gas turbine. Steam turbine perfor-mance is therefore only affected by fuel/feed conditions. Two types of fuel feeding are generally used for FBC’s—dry and wet. For fuels with high heating values the fuel is mixed with water to create a paste (20–25% water). With this method there is naturally no need for coal drying, and evaporated water creates additional mass ﬂ ow through the gas turbine. Dry fuel feeding is more beneﬁ cial with low heating value fuels. FEDERAL AIR EMISSIONS STANDARDS The standards of performance for fossil-fuel-ﬁ red steam gen-erators (constructed after August 17, 1971) were last revised by the federal government as of July 1, 1988. Regulated facilities include fossil-fuel-ﬁ red steam gen-erating units of more than 73 megawatts (heat input rate 250,000,000 Btu/hr.) and fossil-fuel and wood-residue-ﬁ red steam generating units capable of ﬁ ring fossil fuel at a heat input rate of more than 73 megawatts. Existing fossil-fuel-ﬁ red units which have been modiﬁ ed to accommodate the use of combustible materials other than fossil fuels are regulated in a different manner. Within 60 days after the maximum production rate is attained by a regulated facility, the facility must conduct performance tests and provide the E.P.A. with the results of the tests. The tests must also take place before 180 days after the initial start-up of a facility. Each test is speciﬁ c and used for the determination of such things as nitrogen oxide emis-sion. These test methods and procedures may be found in 40 C.F.R. (Code of Federation Regulations) Part 60.46. 6 After a performance text is completed, a facility must not discharge pollutants into the atmosphere at levels greater than those established and listed in the federal regulations. Gases may not contain more than 43 nanograms of par-ticulate matter per joule heat input (0.10 lb. per million Btu) where particulate matter is deﬁ ned as a ﬁ nely divided solid or liquid material, other than uncombined water as measured by the reference methods speciﬁ ed in 40 C.F.R. Part 60.46. These gases must also not exhibit greater than 20% opac-ity except for one six-minute period per hour of not more than 27% opacity. Opacity is deﬁ ned as “the degree to which emissions reduce the transmission of light and obscure the view of an object in the background.” Less stringent standards have been developed for the three following facilities. 6 The Southwestern Public Service Company’s Harrington Station No. 1 in Amarillo, Texas must meet an opacity of not greater than 35%, except that a maximum of 42% opac-ity is permitted for not more than six minutes in any hour. The Interstate Power Company’s Lansing Station Unit No. 4 in Lansing, Iowa must meet an opacity of not greater than 32%, except than a maximum of 39% opacity is permitted for not more than six minutes in any hour. The Omaha Public Power District’s Power Station in Nebraska City, Nebraska must meet an opacity of not greater than 30%, except that a maximum of 37% opacity is permitted for not more than six minutes in any hour. Gases may not contain more than 30 nanograms per joule heat input (0.80 lb. per million Btu) of sulfur dioxide (SO 2 ) derived from liquid fossil fuel or liquid fossil fuel and wood residue. 520 ng/joule heat input (1.2 lb. per million Btu) is the maximum allowable SO 2 discharge from gases derived from solid fossil fuel or solid fossil fuel and wood residue. When different fossil fuels are burned simultaneously in any combination, the SO 2 emission standard is calculated by the following formula: PSso2⫽⫹ ⫹(( ) ( ))/( )y 340 520zyz where PSso2 is the prorated standard in ng/joule heat input derived from all fossil fuels or fossil fuels and wood resi-due ﬁ red, y is the percentage of total heat input derived from liquid fossil fuel, and z is the percentage of total heat input derived from solid fossil fuel. The SO 2 emission standard for Units 1 and 2 at the Central Illinois Public Service Company’s Newton Power Station must comply with the 520 ng/joule requirement if the units individually comply with the 520 ng/joule require-ment or if the combined emission rate from both unites does not exceed 470 ng/joule (1.1 lb/million Btu) combined heat input to both units. It is interesting to note that the federal SO 2 emission limit for West German coal ﬁ red boilers is 2.5 lB./Mbtu (avg.) for boilers of between 18 and 110 MW and 0.51b./MBtu (avg.) for boilers of over 110 MW. 7 Gases may not contain more than 86 ng/joule heat input (0.20 lb/million Btu) of nitrogen dioxide (NO 2 ) derived from gaseous fossil fuel. 129 ng/joule heat input (0.30 lb/million Btu) is the maximum allowable NO 2 discharge from gases derived from liquid fossil fuel, liquid fossil fuel and wood residue, or gaseous fossil fuel and wood residue. 300 ng/joule (0.70 lb/million Btu) is the maximum allow-able NO 2 from solid fossil fuel or solid fossil fuel and wood residue (except lignite or a solid fossil fuel containing 25%, by weight, or more of coal refuse). 260 ng/joule (0.60 lb/million Btu) is the maximum allowable NO 2 from lignite or lignite and wood residue with the exception that 340 ng/C006_001_r03.indd 404C006_001_r03.indd 404 11/18/2005 2:28:15 PM11/18/2005 2:28:15 PM

405joule is the limit for lignite which is mined in North Dakota, South Dakota, or Montana and which is burned in a cyclone ﬁ red unit. When different fossil fuels are burned simultaneously in any combination, the nitrogen oxide emission standard is calculated by the following formula: PSNOxwxyzwxyz⫽⫹⫹ ⫹⫹⫹⫹()()()().260 86 130 300 Where PSNOx is the prorated standard in ng/joule heat input for nitrogen oxides (except nitrous oxide) derived from all fossil fuels or fossil fuels and wood residue ﬁ red, w is the percentage of total heat input derived from lignite, x is the percentage of total heat input derived from gaseous fossil fuel, y is the percentage of total heat input derived from liquid fossil fuel and z is the percentage of total heat input derived from solid fossil fuel (except lignite). There is no standard for nitrogen oxides when burning gaseous, liquid, or solid fossil fuel or wood residue in com-bination with a fossil fuel that contains 25%, by weight, coal refuse. Coal refuse is deﬁ ned as “the waste products of coal mining, cleaning and coal preparation operations (e.g., culm, gob, etc.) containing coal, matrix material, clay, and other organic and inorganic material.” 6 The NO x emission standards for West Germany and Japan are even more stringent than those of the U.S. 7 For new and existing West Germany boilers of over 110 MW, the limit is 0.16 lb./MBtu (6% O 2 ). For Japanese boilers built after 1987, the limit is 0.33 lb./MBtu. PROMINENT FBC INSTALLATIONS IN THE U.S. Recently, in order to reduce SO 2 emissions, Northern States Power Company (NSP) converted its Black Dog pulverized coal-ﬁ red boiler to that of a bubbling bed combustor. This unit is the largest of its kind in the world; its capacity is 130 megawatts. NSP received a new Emissions Permit from the Minnesota Pollution Control Agency (MPCA) for the upgraded unit. The emissions standards set forth in this permit are less strin-gent than those of the federal standards for particulate matter and SO 2 . In the event that utilities should become regulated, the operating parameters of the system or the system itself would have to be modiﬁ ed. 10 The most recent literature available to the author (April 1988) stated that limestone was being added to the bed in order to lower SO 2 emissions sufﬁ ciently to help NSPS stan-dard. The control of particulate matter was difﬁ cult at the onset. However, this problem was resolved by changing the bed material to an inert ﬁ red-clay material. NO x emissions requirements have easily been met. The Tennessee Valley Authority (TVA) has built a 160 MW bubbling bed combustor for the utility’s Shawnee steam plant in Paducah, Kentucky. It has been operating sporadi-cally since autumn of 1988. A pilot plant (20 MW) was completed in 1982 and had brought forth some very promising results. With a Ca : S ratio of 2 to 2.5 (typical range) and a recycle ratio of 2 to 2.5 the SO 2 retention was approximately 90%. 11 This result has been matched by the scaled-up plant. The pilot plant has both an underbed and overbed feed system. Overbed feed does not produce as great a combustion efﬁ ciency as that achieved by the underbed method. This would be expected due to the lack of control over ﬁ nes in the feed. NO x emissions were less than 0.25 lb/million Btu. 11 The NSPS for NO x is 0.7 lb/million Btu for solid fuel. The original underbed feed system was determined to be inadequate because of plugging and erosion problems. The system was redesigned and proved to be successful. The feed system is one of pressurized bottles mentioned earlier in this report under “Bubbling Bed Combustors.” As stated in the “Introduction,” ﬂ uidized bed combus-tion can be used for many different types of industrial pro-cesses. An example of this is the installation of the direct alkali recovery system at Associated Pulp and Paper Mills’ Burnie, Tasmania mill. In this process, sodium carbonate (residual) found in soda-quinone black liquor (a waste product) reacts with ferric oxide to produce sodium ferrite in the combustor (bub-bling bed). The sodium ferrite is then contacted with water to yield sodium hydroxide (desired) and ferric oxide. The ferric oxide is returned to the combustor to be reused. It is interest-ing to note that most of the steam produced in this process is created from the extraction of heat from the exhaust gas and not from bed tubes. The exhaust gas is cleaned via a fabric ﬁ lter and the dust collected is palletized. The pellets are later used in the process. The ﬂ uidizing air is heated from the heat extracted from the hot sodium ferrite after it has been removed from the combustor. Since there is no sulfur involved in this process the exhaust gas is easily cooled, thereby allowing greater pro-duction of high-pressure steam. 13 The title for the world’s largest CFB probably belongs to the nuclear generating station owned by Colorado-Ute Electric Association. The original 25-year-old plant was replaced because it was uneconomical to operate. The capac-ity of the new plant is 110 MW. In May of 1988 on EPRI (Electric Power Research Institute) assessment began and is scheduled to continue until May of 1990. As of April 1988, the unit was reported to be easy to operate, responsive to load variations, and easily restarted following a trip. However in 1989 opera-tional difﬁ culties were reported. SO 2 emissions standards were expected to be easily met and NO x emissions were well under the limit. Final determination of the optimum Ca : S ratio still needed to be determined. Particulate matter emissions are expected to be less than 0.03 lbs/million. Btu because of the addition a new baghouse to the existing three baghouses. Some valuable information has been learned from the unit thus far, e.g., control of coal feed size has been impor-tant in maintaining the bed quality and agglomerations can be avoided if the feed is started in short bursts prior to being C006_001_r03.indd 405C006_001_r03.indd 405 11/18/2005 2:28:15 PM11/18/2005 2:28:15 PM
406
continuous; this is to allow the temperature rise to be more uniform. 14 One of the larger commercial units in the U.S. is located in Colton, California and was installed for Cal-Mat Co. The 25 MW CFB was constructed because electric utility rates were rising and the availability of power was uncertain. The company manufactures cement—a process requiring much electricity. Since the company had easy access to coal and limestone as well as a large quantity of heat from its kilns, CFB technology became an effective solution to their energy needs. Bottom ash waste and ﬂ yash could also be used in the cement-making process. As could be expected, the air pollution controls instituted by the state of California are very strict. However, a permit was granted to CalMat in a relatively short period of time because of the ﬁ ne performance demonstrated by this unit. The exhaust gases were found to contain SO 2 at 30 lb/hr., NO x at 57 lb/hr. and CO at 24 lb/hr. 15 There were initial problems with equipment and systems, however, these were eventually eliminated. Bed retention and temperature control problems have also been resolved through modiﬁ cations of the air ﬂ ows and nozzles. PFBC’s have been installed in Sweden, the U.S. and Spain. Two PFBC modules of 200 MW each have been installed in Vartan, Stockholm. The ﬁ rst unit is due for start-up in late 1989. The Swedish emission standards are very strict and include special restrictions on noise and dust since the units are located very close to a residential area. A 200 MW combined cycle PFBC will be installed by American Electric Power (AEP) at its Tidd Power Plant at Brilliant, Ohio. Test results from joint studies proved PFBC technology to have environmental beneﬁ ts surpassing those of traditional boilers with ﬂ ue gas desulfurization systems (FGD), selective catalytic reduction, etc. A 200 MW PFBC will be installed by Empresa Nacional de Electricidad S.A. (ENDESA) at its Escatron Power Plant as a retroﬁ t for an existing unit. 90% sulfur removal and a NO x decrease of 30% are expected. Many different technolo-gies were considered but PFBC was chosen because of the high sulfur/ash/moisture black lignite coal that they burn. 16 NO X /SO 2 FORMATION AND CONTROL Fossil fuels naturally contain sulfur in varying percentages. As fuel is burned the sulfur combines with oxygen to form SO x , and primarily, SO 2 . When emitted into the atmosphere this SO 2 can combined with water vapor to form sulfuric acid (and sulfurous acid to a lesser extent). This is a part of the basic mechanism by which acid rain is created. In order to control sulfur dioxide emissions, the oldest and still most common method used is to react the gas with limestone or a similar calcium based material. Crushed lime-stone (CaCO 3 ) can be fed continuously to a conventional coal boiler or ﬂ uidized bed where it calcines to lime (CaO) and then reacts with SO 2 in the presence of oxygen to form calcium sulphate (CaSO 4 ). This material precipitates to the bottom of the combustor and is removed. Coal particle size has a deﬁ nite impact on desulfuriza-tion. Bed composition also has an effect on sulfur removal. A typical bed might be composed of coarse partially sul-fated limestone and ash (produced by combustion). The par-ticle size of the coal and limestone would probably be equal however combustor operation conditions such as ﬂ uidizing velocity will dictate the particle size. An alternate scenario might be to pulverize the limestone and introduce it to a bed composed of ash or some other type of refractory material. Fines naturally have shorter residence times than do coarse materials and, hence, would probably have to be recycled to increase efﬁ ciency. A series of experiments were carried out by Argonne National Laboratory 17 using three different types of lime-stones to test their effects on sulfur capture during com-bustion. The average particle size of the limestone was 500–600 micrometers. The Ca : S ratio was 2.3–2.6 and the combustion temperature was 1600⬚F. SO 2 removal was 74 to 86%. The test proved that the amount of SO 2 removal was relatively independent of the type of limestone used. The test also proved that particle size did not have much of an effect on SO 2 removal. The explanation offered for this observation was that although larger particles are less reac-tive than smaller particles, the increased residence time in the combustor of larger particles compensates for the lower reactivity. Dolomite was also evaluated for SO 2 capture. In two experiments, Tymochtee dolomite was added to a bed composed of alumina at Ca : S ratios of 1.5 and 1.6. The average particle size was 650 micrometers. The SO 2 remov-als were 78% and 87% respectively. MgO is contained within the dolomite matrix and is believed to keep the par-ticles more porous such that sulfation is greater, especially in larger particles. Combustion temperature had a marked effect on SO 2 removal in these experiments. Dolomite No. 1337 was most effective in reducing SO 2 at 1480⬚F. Limestone No. 1359 was most effective in the range of 1500–1550⬚F. Both sorbents achieved approximately 91% SO 2 removal. The average particle size was approximately 500 micrometers. Pulverized limestone No. 1359 with an average particle size of 25 micrometers was most effective in the range of 1550–1600⬚F. The extent of calcination is more dependent upon bed temperature for ﬁ nely pulverized limestone. The greater the calcination, the greater the reactivity with SO 2 . An unusual ﬁ nding occurred in that Tymochtee dolomite was observed to be most effective in SO 2 removal at 1800⬚F. For all of the other sorbents the SO 2 removal was very poor at this temperature. Explanations for this phenomenon have been proposed. One explanation suggests that above a certain temperature the sorbent’s pores close thereby ending sulfona-tion. Depending upon the sorbent’s structure and composition, this temperature would be different for each sorbent. Another explanation involves the effect of ﬂ uidized bed gas circulation on bed chemistry. An emulsion phase and a gas bubble phase C006_001_r03.indd 406C006_001_r03.indd 406 11/18/2005 2:28:15 PM11/18/2005 2:28:15 PM

407exist in any gas-solid ﬂ uidized bed. Excess gas which is not needed for ﬂ uidization circulates back and forth between the two phases. This gas does not react with the sorbent until it reaches the emulsion phase. All of the oxygen in the lower portion of the emulsion phase reacts with the fuel to from CO. As the bubbles rise through the bed, air exchanges between the bubbles and the emulsion phase. The upper portion of the emulsion phase contains excess oxygen. The following reac-tion was thus proposed as one which takes place in the lower portion of the combustor: CaSO CO CaS CO42⫹⫹4 U 4 . (1) One or more of the following reactions were proposed to occur in the upper portion of the combustor: CaS 1 O CaO SO22⫹⫹12U (2) CaS CaO SO2⫹344CaSO4U ⫹ (3) CaS CaSO4⫹ 2O2U . (4) Reactions 2 and 3 would limit the unit’s ability to remove sulfur because of the regeneration of SO 2 . This regeneration of SO 2 is so dependent upon temperature that it could very possibly be an explanation as to why SO 2 removal generally suffers at high temperatures. Nitrogen also occurs naturally in fossil fuels. This nitro-gen reacts with oxygen during combustion and later forms acid rain in very much the same manner as with sulfur. Oxides of nitrogen (NO x ) are also responsible for the cre-ation of “smog.” As nitrogen dioxide (NO 2 ) absorbs light of certain wavelengths it dissociates photochemically to form nitric oxide (NO) and atomic oxygen. This atomic oxygen is very reactive and readily combines with O 2 to from ozone (O 3 ). Ozone in turn oxidizes hydrocarbons in the air to form aldehydes. Ozone and the aldehydes are components of smog. NO 2 is the reddish-brown gas which can often be seen on the horizons of cities such as Los Angeles. The principal oxide of nitrogen formed during com-bustion is nitric oxide. Nitrogen in the fuel combines with oxygen in the ﬂ uidizing air as follows: 1212NONO.22⫹ U The kinetics of NO decomposition are slow enough so that equilibrium levels are not achieved. Various experiments conducted by Argonne National Laboratory as well as by other researchers have proven that most of the nitrogen forming NO x is from the fuel and not from the air. This has been easily demonstrated by substituting an inert gas (such as argon) for nitrogen in the ﬂ uidizing air and then comparing the results to those of combustion with standard ﬂ uidizing air. As previously mentioned in this report two-stage com-bustion is an effective method of decreasing NO x emissions. As with SO 2 reduction bed composition has an important effect on NO x . It has been determined through experimenta-tion and experience that limestone also decreases NO x emis-sions. Skopp and Hammons 18 observed that when using a limestone bed two factors were changing with time which could have been responsible for decreasing NO emissions: the CaSO 4 concentration in the bed was increasing and so was the SO 2 concentration. The increase in CaSO 4 sug-gested that it could be a selective catalyst for reduction of NO. The increasing SO 2 concentration suggested that there might be a reaction occurring between it and the NO which was lowering the NO. This was investigated by conduct-ing experiments using synthetic NO—SO 2 —N 2 gas mix-tures. The results showed that no reaction in the gas phase occurred. There was also no reaction between the NO and SO 2 over CaSO 4 . However, there was a reaction occurring over a bed of 20% sulfated lime. This reaction was found to have a negative temperature dependence. The following mechanism was proposed by Skopp and Hammons 18 as an explanation for their results: CaO SO CaSO23⫹ U CaSO NO CaSO (NO332⫹ 2 U ) CaSO (NO CaSO N O32 42) U ⫹ NO N O222U ⫹12. Esso researchers investigated the possibility of NO being produced by CO catalyzed by CaSO 4 . The rate of this reac-tion was found to increase with increasing temperature. Argonne researchers 17 investigated the use of metal oxides, among them, aluminum oxide (Al 2 O 3 ), zirconium oxide (ZrO 2 ) and cobalt oxide (Co 3 O 4 ). At the time these experiments were conducted, the literature had indicated that these metal oxides were effective in reducing or catalytically decomposing NO. The results showed that the addition of Al 2 O 3 and ZrO 2 did nothing to reduce NO formation during combustion in a ﬂ uidized bed. The addition of Co 3 O 4 actu-ally increased rather than decreased the formation of NO. A study was conducted by McCandless and Hodgson 20 for the U.S.E.P.A. on the use of metal sulﬁ des as a way to reduce NO emissions. The following is well known as the “Thiogen” process and has been used in the recovery of sulfur from SO 2 CaS 2SO CaSO S242⫹⫹U4CaS SO CaSO S232⫹⫹64 3U . Based on this process it was determined that the following reaction might also be possible CaS 4NO CaSO 2N .42⫹⫹U C006_001_r03.indd 407C006_001_r03.indd 407 11/18/2005 2:28:15 PM11/18/2005 2:28:15 PM
408
Preliminary studies indicated that the reaction did pro-ceed and could be an effective method for NO x control. Nineteen metal sulﬁ des were used. All but one reduced NO to N 2 at temperatures between 194⬚F and 1202⬚F. However, a weight loss did occur indicating that an undesirable side reaction was taking place—probably the formation of SO 2 . Some metal SO 4 was formed in most of the tests. However, the alkaline earth sulﬁ des were determined to be the most stable. It was also found that the temperature at which the reduc-tion reaction occurs can be lowered if certain catalysts such as NaF and FeCl 2 are mixed with the sulﬁ des. Reaction tem-perature was again reduced when the sulﬁ de/catalyst combi-nation was impregnated on alumina pellets. Tests were also conducted involving synthetic ﬂ ue gas containing 1000 ppm NO, 1% O 2 , 18% CO 2 and the remainder N 2 . Using this gas in combination with the CaS showed that NO was signiﬁ -cantly reduced above temperatures of 1112⬚F, by using the sulﬁ de/catalyst combination. The results of the experiments showed that between 0.372 and 0.134 grams of NO were reduced per gram of metal sulﬁ de. Between 0.76 and 0.91 grams was achieved when using the impregnated alumina pellets. The authors recommended that more research be done to evaluate the economical implications of using these materials. Several other interesting facts known about NO x con-trol and found in the literature are that increasing ﬂ uidizing velocity decreases NO x , NO x is not signiﬁ cantly affected by excess air, and NO x production increases at lower tempera-ture, especially below approximately 1500⬚F. For conventional coal-ﬁ red boilers the most common approach to control NO x and SO x simultaneously is the combination of selective catalytic reduction (SCR) and wet-limestone or spray dryer ﬂ ue gas desulfurization (FGD). The SCR process converts NO x to N 2 and H 2 O by using ammo-nia as a reducing agent in the presence of a catalyst. The catalytic reactor is located upstream from the air heater and speeds up the reaction between the NO x and the ammonia, which is injected into the ﬂ ue gas in vapor form immediately prior to entering the reactor. The reduction reactions are as follows: 21 44 46NH NO O N H O3222⫹⫹ ⫹U 42 36NH NO O N H O.32222⫹⫹ ⫹U It can be seen that the amount of NO 2 removed primarily depends on the amount of NH 3 used. Although SCR technol-ogy has proved to be an effective means to reduce NO x with removal results as high as 90% in some European facilities, the U.S. does not consider the technology economically fea-sible. In addition to the high cost there are the undesirable effects of unreacted NH 3 , by-product SO 3 and increased CO production to consider. There are also catalyst deactivation problems caused by contamination by trace metals in the ﬂ y ash and by sulfur poisoning. The Japanese have improved on the design of catalysts and their arrangement within the reactor. However, these modiﬁ cations are still too new to evaluate their merit. 22 U.S. industry also feels that more data has to be generated for the medium to high sulfur coals most commonly used in this country. Since characteristics such as high sulfur, low ﬂ yash alkalinity and high iron content are common in U.S. coal, and these qualities do inﬂ uence SO 3 production, SCR would not appear to be one of the likely options for U.S. industry at least in the near future. Exxon has developed a process called “Thermal deNO x ” which makes use of ammonia injection into the ﬂ ue gas at temperatures of between 1600 and 2200⬚F. This process is claimed to remove NO x by up to 90%. years, CFB’s have become the dominant FBC choice in industry. The most common problems that have been associated with bubbling beds include erosion of the inbed tubes. This can be reduced through the use of studding, ﬁ ns, etc. as previously men-tioned in this report. However, CFB’s are also prone to ero-sion, i.e., the waterwalls, as well as the refractory lining. Agglomeration is another common problem associated with bubbling beds. Sand can fuse in localized hot spots to form clinkers or “sand babies” especially when the fuel has a high concentration of alkali compounds. 9 In severe cases, agglomerations can cause the bed to deﬂ uidize, block air ports, and make bed material removal more difﬁ cult. Sulfur removal is more difﬁ cult with bubbling beds. In general, large quantities of double-screen stoker coal must be used to attain the high sulfur removal rate displayed by CFB’s. Most overbed feed bubbling beds in existence must use coals which contain less than 10% ﬁ nes. This can often be quite costly. As previously mentioned, underbed feed also has problems associated with it. Since low ﬂ uidizing veloci-ties are required with underbed feed, the bed plan area must be larger and, subsequently, contain a higher density of feed ports. This serves to complicate the already unreliable feed system. In order to utilize the sorbent better, the recycle ratio has to be increased. However, above a certain recycle ratio, and in-bed tubes might have to be removed in order to maintain combustor temperature, compromising the CFB design. NO x control is better with CFB’s than with bubbling beds. This is because of the aforementioned stage combus-tion which is physically unachievable in bubbling beds due to the large bed plan area and low ﬂ uidizing velocity. On average, 0.1 lb/million Btu less NO x is produced by CFB’s than by bubbling beds. As of the present, there are no federal regulations gov-erning CO emissions. However, some states have promul-gated regulations. As would be expected with overbed feed bubbling bed combustors, the CO emissions are high. While emissions of over 40 ppmv are common with bubbling beds, CFB’s are usually under 100 ppmv. 3 This is due to better circulation and recycle. There is not much data on CO emis-sions for underbed feed bubbling beds. However, it evidently reduces CO more than does overbed. Unfortunately, with CFB’s there is a trade-off between SO 2 /NO x and CO. Staged combustion will increase CO emissions as the primary to secondary air ratio becomes smaller. SCR/SNR speciﬁ c to CO also may cause an increase in NO x . C006_001_r03.indd 408C006_001_r03.indd 408 11/18/2005 2:28:15 PM11/18/2005 2:28:15 PM

409 Other problems associated with bubbling beds are scale-up and turn-down. Scale-up is limited because of the feed distribution problem and turn-down is usually more frequent because of the erosion problem. Incidentally, it should have been mentioned before, that, in general, FBC’s take longer to start up and turn down than conventional boilers because of the large amounts of bed material which must be heated or cooled. The major advantage which PFBC’s hold over AFBC’s is that NO x , SO 2 , and CO are weakly linked. Thus, 60 mg/MHJ NO x can be attained while at the same time only 50 mg/MJ SO 2 (or less) and 10 mg/MJ (or less) CO are produced. Another advantage of PFBC’s is that the waste contains negligible lime, sulﬁ des, and sulﬁ tes. The decreased lime concentration makes the waste less reactive and probably renders it nonhazardous. The decreased sulﬁ tes makes the Concentration (wt percent)Component Test no.1aText no. 2aTest no. 3aBaghousebFull-scale residuesCaSO426.1 31.6 21.7 27.7 26.1CaS 0.8 0.6 0.7 5.2 0.45Free CaO 24.8 28.3 24.7 15.7 23.5CaCO33.4 3.2 5.2 6.8 4.6Fe2O310.7 9.5 9.1 10.6 15.9Other mainly SiO2 and C 27.3 20.8 33.8 19.5 24.2LOIc (corrected)11.4 5.4 5.1 8.0 4.3Sum 104.5 99.4 100.3 — 99.0 a Composite pilot-scale residues. b Calculated from TGA and other analyses. c LOI indicated loss on ignition. TABLE 1 Major chemical components of composite residues: pilot and full-scale CFBC units Pilot-scale composite residuesPilot-scale baghouse residue Full-scale residueSpecific gravity 2.83–3.07 2.58 2.95Mean size, D50 mm 0.2 0.04 0.04Optimum water content percenta 14.5–17.5 32 26.5–30.5Unconfined compressive strength, kPabCuring period, days0 230–360 — —3 — — 24707 150–290 — 41208 — 309 —10 260–425 385 —12 — 461 —28 — — 4660Freeze/thaw cycles4 — 201 —6 540–1020 Sample destroyed —7 — — 320015 — — 880 a As determined by standard Proctor test. b Samples were cured at 100 percent relative humidity at 23 ± 2⬚C for periods shown. TABLE 2Geotechnical properties of the pilot-scale rig composite and baghouse samples and full-scale unit residuesC006_001_r03.indd 409C006_001_r03.indd 409 11/18/2005 2:28:16 PM11/18/2005 2:28:16 PM
410
material a good candidate for use in cement kilns and con-crete. The maximum sulfur content per ASTM standards is 1.2% by weight or 3.1% by weight as sulﬁ tes. The material has been found useful for building roads, manufacturing gravel and formed bricks or tiles, and for rooﬁ ng and ﬂ ooring material. 3 Other advantages of PFBC’s include compactness because of smaller bed requirements, plant cycle efﬁ cien-cies of 40–42% and subsequent reduced fuel costs, and unit modularity for ease in increasing future capacity. 8 Disadvantages include in-bed tube erosion and potential damage to the gas turbine if the hot gas clean-up is ineffec-tive. It should be noted that the technology is too new to accurately assess its advantages versus its disadvantages. CHARACTERIZATION OF SOLID WASTES FROM FLUIDIZED BEDS The characterization and use of ﬂ uidized-bed-combus-tion coal/limestone ash is discussed in the articles of Behr-Andres and Hutzler 23 and Anthony et al. 24 The former dealt with the use of the mixture in concrete and asphalt. The latter presented chemical and physical properties for the waste (see Tables 1 and 2 above). Hot-gas cleanup (HGCU) technologies have emerged as key components of advanced power generation technologies such as pressurized ﬂ uidized-bed combustion (PFBC), and integrated gasiﬁ cation combined cycle (IGCC). The main difference between HGCUs and conventional particulate removal technologies (ESP and baghouses) is that HGCUs operate at higher temperatures (500 to 1,000⬚C) and pres-sures (1 to 2 MPa), which eliminates the need for cooling of the gas. See Website (2005): http://www.worldbank.org/html/fpd/em/power/EA/mitigatn/aqpchgas.stm REFERENCES 1. Robert H. Melvin and Reid E. Bicknell, “Startup and Preliminary Operation of the Largest Circulating Fluid-Bed Combustion Boiler in a Utility Environment—NUCLA CFB Demonstration Project,” Paper Presented at the 50th American Power Conference, Chicago, Illinois, April 18–20, 1988 p. 2. 2. Jason Makansi and Robert Schweiger, “Fluidized-bed boilers,” Power, May 1987, p. 9. 3. Efficiency and Emissions Improvements by Means of PFBC Retrofits (Finspong, Sweden: Asea PFBC Component Test Facility, S-61220, 1988), p. 2. 4. Taylor Moore, “Fluidized bed at TVA,” EPRI Journal, March 1989, p. 27. 5. Asea Babcock PFBC Update, 1, No. 3 (Fall 1988), n. pag. 6. Code of Federal Regulations, Vol. 40, Part 60, Revised as of July 1988. 7. Charles Sedman and William Ellison, “German FGD/DeNO x Expe-rience,” Presented at the Third Annual Pittsburgh Coal Conference, Pittrsburgh, Pennsylvania, September 1986. 8. Asea Babcock PFBC Update, 1, No. 2 (Summer 1988), n. pag. 9. Jason Makansi, “Users pause, designers wrestle with fluid-bed boiler scaleup,” Power, July 1988, p. 2. 10. David Osthus, John Larva, and Don Rens, “Update of the Black Dog Atmospheric Fluidized-Bed Combustion Project,” Paper Presented at the 50th American Power Conference, Chicago, Illinois, April 18–20, 1988, p. 1. 11. Bob Schweiger, ed., “Fluidized-bed boilers achieve commercial status worldwide,” Power, Feb. 1985, p. 9. 12. R.A. Cochran and D.L. Martin, “Comparison and Assessment of Cur-rent Major Power Generation Alternatives,” Presented at the Power-Gen Exhibition and Conference for Fossil and Solid Fuel Power Generation in Orlando, Florida, Boston, Massachusetts, Dec. 1988, p. 2. 13. Sheldon D. Strauss, “Fluidized bed keys direct alkali recovery,” Power, Feb. 1985, p. 1. 14. Melvin and Bicknell, p. 6 (see 1). 15. Bob Schweiger, ed., “U.S.’s largest commercial CFB burns coal cleanly in California,” Power, Oct. 1986, p. 2. 16. Efficiency and Emissions Improvement by Means of PFBC Retrofits, p. 10. 17. A.A. Jonke et al. , “Reduction of Atmospheric Pollution by the Appli-cation of Fluidized-Bed Combustion,” Argonne National Laboratory, Publication No. ANL/ES-CEN-1002, 1970, n. pag. 18. A. Skopp and G. Hammons, “NO x Formation and Control in Fluidized-Bed Coal Combustion Processes,” ASME Winter Annual Meeting, Nov./Dec., 1971. 19. Jason Makansi, “Meeting future NO x caps goes beyond furnace modi-fications,” Power, September 1985, p. 1. 20. Reduction of Nitric Oxide with Metal Sulfides, Research Triangle Park: U.S. E.P.A., EPA-600/7078–213, Nov. 1978, pp. 1–5. 21. Ibid, p. 3. 22. Ed Cichanowicz, “Selective catalytic reduction controls NO x in Europe,” Power, August 1988, p. 2. 23. Christina B. Behr-Andres and Neil J. Hutzler, “Characterization and use of fluidized-bed-combustion coal ash,” Journal of Environmental Engineering, November/December 1994, p. 1488–506. 24. E.J. Anthony, G.G. Ross, E.E. Berry, R.T. Hemings. and R.K. Kissel, “Characterization of Solid Wastes from Circulating Fluidized Bed Combustion”, Trans. of the ASME Vol. 117, March 1995, 18–23. JAMES SANDERSON Environmental Protection Agency Washington, D.C. FISH ECOLOGY: see POLLUTION EFFECTS ON FISH; THERMAL EFFECTS ON FISH ECOLOGY C006_001_r03.indd 410C006_001_r03.indd 410 11/18/2005 2:28:16 PM11/18/2005 2:28:16 PM
411

bundet

The amount of pollutants, especially sulfur oxides and par-ticulates emitted to the atmosphere may be reduced by treat-ing fuels prior to combustion. This approach may be more energy efﬁ cient than treatment of ﬂ ue gases as per Vapor and Gaseous Pollutant Fundamentals. More than thirty million tons of sulfur dioxide are discharged annually in the United States, 75% of which is the result of fuel burning. FOSSIL FUEL PRODUCTION, RESERVES AND CONSUMPTION The world’s production of oil in 1980 was 66 million barrels per day with a projected value of 77 MBPD for the year 2000. The relatively small anticipated increase reﬂ ects increased con-servation and alternate fuel source application. The overall oil output of the USSR was about 14 MBPD 1 as compared to about 12 MBPD for combined US and Canadian production (1980). About 2500 trillion cubic feet of natural gas reserves are estimated to exist worldwide. The US reserves are 200 TCF with an annual consumption of about 20 TCF. Soviet bloc production was about 15 TCF in 1980. Most oil and natural gas reserves fall in a crescent shaped area extending from Northern Algeria Northward to West Siberia. Lynch 2 felt that the level of surplus capacity would remain stable for the early ’90s with the then world stock level of about 100 gigaliters (1.3 giga barrels). Coal is consumed at a rate of 600 million tons annually in the US utility industry. Only a small portion of Eastern US coals fall in the low (less than 1% sulfur) category—see Table 1. The US, USSR and China own about two thirds of the world’s 780 billion tons of presently recoverable coal reserves. The US has about one quarter of the total. Coal accounts for 90% of the US’s proven reserves. 3 Consumption of fuel might be measured in “quads” or quadrillion Btu’s. It has been estimated that US electric con-sumption was 13 quads and nonelectric industrial about 16 quads for the year 1980. 3 Total US fossil fuel consumption is about 76 quads, most in the non-industrial sector. Worldwide energy consumption is predicted to double over the next 25 years according to the World Energy Council. 3a The pre-dicted fossil fuel usage in terms of billions kwh electric gen-eration in the year 2015 is for coal-2000, natural gas-1000, nuclear-400, and petroleum-less than 100. Renewables are estimated at 400 billion kwhs. Divide these numbers by 100 to estimate the number of quads; assuming a plant efﬁ ciency of current Rankine cycle plants (about 34%) or by 170 if a combined cycle (Brayton ⫽ Rankine) is assumed. SULFUR REMOVAL Typical legislative actions have been the setting of limits on the allowable sulfur content of the fossil fuel being burned or on the SO 2 emission rates of new sources. In California, regula-tions have limited the use of fuel oil to those of 0.5% or less sulfur. Since 1968, a limit of a 0.3% sulfur oil has been in effect in New York City. In 1980, Massachusetts set a 1% sulfur limit on the coal to be burned. This limit is being considered for other Atlantic seaboard states as coal conversion is increasingly encouraged. Chemical and physical desulfurization of fossil fuels can be used to produce levels of sulfur which comply with government standards. To reduce a 3% sulfur coal to a 1% sulfur coal may add about 10% to the cost of coal F.O.B., but may save on transportation and ﬂ ue gas desulfurization costs. The amount of sulfur dioxide emitted worldwide might double in the next decade due to increased energy demands (approximately 3.5% annually) and the use of more remote crudes having higher sulfur concentration. The chemical and power industries must strike a delicate balance between the public’s dual requirement of increased quantities and preparation of fossil fuel. More fuel must now be desulfurized more completely and/or more sulfur diox-ide must be removed from stack gases. The techniques for cleaning fossil fuels used throughout the petroleum, natural gas and coal production industries are covered in this arti-cle. Treatment of stack gases to effect particulate and sulfur removal are discussed separately in other articles. PROCESSES INVOLVING THE BASIC FUELS The two most commonly combusted energy sources are coal and fuel oil having typical sulfur ranges of 1–4% and 3–4%, respectively; a 3% sulfur oil produces about the same SO 2 emission as a 2% sulfur coal when based on a comparable energy release. Fuel oil desulfurization is used by most major oil producers. Hydrogenation, solvent extraction, absorption and chemical reaction are used to varying extents at petro-leum reﬁ ners. Finfer 4 claims a possible sulfur reduction from 2.5 to 0.5% by a hydrodesulfurization process. Coal contains sulfur which may be combined with either the organic or C006_002_r03.indd 411C006_002_r03.indd 411 11/18/2005 10:27:08 AM11/18/2005 10:27:08 AM
412
inorganic (pyritic and sulfate) matter. The organics may be removed by various cleaning processes, but little reduction in organic sulfur has been found to occur by physical clean-ing methods. Currently an extraction process, followed by hydrogenation, is being tried. Some coals have been reduced to S contents below 2%, and typical sulfur reduction esti-mates are in the range of 20–40% reduction. 5,6,7 Even if these reduced levels are achieved, a need for further removal of sulfur from the ﬂ ue gases might exist. Cleaning, when com-bined with ﬂ ue gas desulfurization as a method of SO 2 con-trol, could eliminate the need for reheat and considerably reduce the sludge handling requirements of the plant. Fuel Oil Desulfurization (General) Before the ecological need for fuel oil desulfurization was recognized, oil stocks were desulfurized for a number of other reasons: 1) To avoid poisoning and deactivation of platinum cat-alysts used in most catalytic reforming processes. 2) To reduce sulfurous acid corrosion of home burner heating equipment. 3) To demetalize crude stocks (sulfur removal from crude is generally accompanied by a concomitant removal of such trace metals as sodium, vanadium and nickel). 4) To recover pure sulfur. 5) To reduce or eliminate final product odor. By deﬁ nition, hydrodesulfurization is the removal of sulfur by a catalytic reaction with hydrogen to form hydrogen sulﬁ de. As carried out in the petroleum industry, the hydrode-sulfurization process is not a speciﬁ c chemical reaction. Various types of sulfur compounds (mercaptans, sulﬁ des, polysulﬁ des, thiophenes) with varying structures and molecu-lar weights are treated. Obviously, they react at various rates. TABLE 1Ash content and ash fusion temperatures of some U.S. coals and ligniteRankLow Volatile Bituminous High Volatile Bituminous Subbituminous LigniteSeam Pocahontals No. 3 No. 9 Pittsburgh No. 6Location West Virginia Ohio West Virginia Illinois Utah Wyoming TexasAsh, dry basis, % 12.3 14.10 10.87 17.36 6.6 6.6 12.8Sulfur, dry basis, % 0.7 3.30 3.53 4.17 0.5 1.0 1.1Analysis of ash, % by wt———————SiO260.0 47.27 37.64 47.52 48.0 24.0 41.8 Al2O330.0 22.96 20.11 17.87 11.5 20.0 13.6 TiO21.6 1.00 0.81 0.78 0.6 0.7 1.5 Fe2O34.0 22.81 29.28 20.13 7.0 11.0 6.6 CaO 0.6 1.30 4.25 5.75 25.0 26.0 17.6 MgO 0.6 0.85 1.25 1.02 4.0 4.0 2.5 Na2O 0.5 0.28 0.80 0.36 1.2 0.2 0.6 K2O 1.5 1.97 1.60 1.77 0.2 0.5 0.1 Total 98.8 98.44 95.74 95.20 97.5 86.4 84.3Ash fusibility — — — — — — —Initial deformation temperature, F Reducing 200+ 2030 2020 2000 2060 1990 1975 Oxidizing 2900+ 2420 2265 2300 2120 2190 2070Softening temperature, F Reducing 2450 2175 2160 2180 2130 Oxidizing 2605 2385 2430 2220 2190Hemispherical temperature, F Reducing 2480 2225 2180 2140 2250 2150 Oxidizing 2620 2450 2450 2220 2340 2210Fluid temperature, F Reducing 2620 2370 2320 2250 2290 2240 Oxidizing 2670 2540 2610 2460 2300 2290C006_002_r03.indd 412C006_002_r03.indd 412 11/18/2005 10:27:09 AM11/18/2005 10:27:09 AM

413In addition, during the course of desulfurization, non-sulfur containing molecules may be hydrogenated and in some cases cracked. The ﬂ ow design of hydrosulfurization process systems is relatively simple. Preheated oil and hydrogen under pres-sure are contacted with catalyst. The efﬂ uent from the reac-tor is passed to one or more separators to remove most of the efﬂ uent hydrogen and light hydrocarbon gases produced in the operation. These gases are generally recycles with or without prior removal of light hydrocarbons by absorption. The separa-tor liquids may be stripped, rerun or otherwise treated to obtain hydrogen sulﬁ de free products of the desired boiling range. Except in the case of residuum processing, plant design options are few in number and relatively simple. For exam-ple, in the processing of distillates, correlation systems have been developed which relate degree of desulfurization to about three parameters which deﬁ ne the charge stock, reac-tor temperature, temperature, pressure, feed space velocity, hydrogen rate and a catalyst activity parameter. When residuum stocks are considered, however, general-izations are not so easily made. The wide variance in resid-uum properties (i.e., atmospheric or vacuum type, viscosity, Conradson carbon content, metal content and the parafﬁ nic or aromatic nature of residuum) makes each case a special one as far as process design. Catalyst poisoning due to metals deposition on the catalyst surface can reduce overall desulfur-ization yields. Catalyst must then be regenerated or replaced, thus adding to overall cost of the particular system employed. An alternative to desulfurization exists, that being the use of natural low sulfur fuel oils. They may be used alone or in blends with higher sulfur content material. The major source of low sulfur fuel oil is North African crudes, princi-pally from Libya and Nigeria, and some Far Eastern crude from Sumatra. Fuels made from these crudes will meet even very low sulfur regulations calling for 0.5% sulfur or less. However, the highly waxy nature of these parafﬁ nic materi-als makes handling difﬁ cult and costly. Therefore, the blend becomes a more palatable course of action. Blends of natural low-sulfur fuels oils with other high sulfur fuel oils will be adequate in some cases to meet more moderate sulfur regulations. The fuel oil fractions of North African crudes contain about 0.3% S. Thus signiﬁ -cant amounts of higher sulfur fuel oils can be added to make blends calling for 1–2% sulfur. These blends have physical properties which obviate the need for specialized handling (a must for existing industrial installations). Before delving into speciﬁ c desulfurization technology and applications, pertinent terms will be deﬁ ned. Figure 1 FUEL OILNAPHTHALENNO. 6 FUEL OILBENZENETOLUENEHGIFELATM GAS OILH2KEROSENEPREMIUM GASOLINERES. GASOLINEBUTANENAT GASODCNAPHTHAABCRUDELIGHT ENDSLIGHT REFORMATEHEAVY REFORMATEGENERAL FLOWSHEET - CRUDEOIL PROCESSINGLEGENDA - CRUDE DISTILLATIONB - CATALYTIC REFORMERC - BTX EXTRACTIOND - GASOLINE POOLE - PYROLYSISFGHI - HYDRODEALKYLATIONHYDROTREATERTARALKYL NAPHTHALENEFIGURE 1C006_002_r03.indd 413C006_002_r03.indd 413 11/18/2005 10:27:09 AM11/18/2005 10:27:09 AM
414
schematically represents a general ﬂ owsheet for crude oil pro-cessing. Crude oil, as received from the source is ﬁ rst atmo-spherically distilled. Light ends and mid-distillates from this operation are further processed to yield gasolines and kero-sene. Atmospheric residuum can be directly used for No. 6 fuel oil, or further fractionated ( in vacuo ) to produce vacuum gas oil (vacuum distillate) and vacuum residuum. After atmo-spheric distillation, the average crude contains about 50% of atmospheric tower bottoms, which is nominally a 650⬚F ⫹ oil. The vacuum distillation yields roughly equal parts of vacuum gas oil and vacuum residuum. The bottoms from this unit is nominally a 975⬚F ⫹ oil, although the exact cut point will vary for each vacuum unit. Desulfurization of vacuum residuum would be appli-cable where a reﬁ nery has use for the virgin vacuum gas oil other than fuel oil, and sulfur restrictions or increased prices make desulfurization of vacuum bottoms attractive. Another situation is where desulfurizing the vacuum gas oil and blending back with vacuum bottoms no longer produces a ﬁ nal fuel oil meeting the current sulfur speciﬁ cation. Present in the residuum (vacuum) is a fraction known as asphaltenes. This portion is characterized by a molecu-lar weight of several thousand. The majority of the organo-metallic compounds are concentrated in the asphaltene fraction. Although many of the metals in the periodic table are found in trace quantities, vanadium and nickel are usu-ally present in by far the highest amounts. Residual oils from various crudes differ from each other considerably in regard to hydrodesulfurization. These differences reside to a great extent in the asphaltene fraction. Light Oil Desulfurization The G O-Fining Process The G O-Fining process is designed for relatively complete desulfurization of vacuum gas oils, thermal and catalytic cycle oils, and coker gas oil. It represents an extremely attractive alternative where a lesser degree of sulfur removal from the fuel oil pool and/or a very low sulfur blending stock is required. The feed to the G O-Finer System is atmospheric residuum. This stream is vacuum fractionated and the resulting vacuum gas oil (VGO) is desulfurized using a ﬁ xed bed reactor system. Resultant VGO is then reblended with vacuum bottoms to yield a desulfurized fuel oil or used directly for other applications. Figure 2 shows quantitative breakdown of various process streams for a 50,000 barrel per stream day (BPSD) operation utilizing a 3% sulfur Middle East atmospheric residuum feed. The process has the capa-bility of producing 49,700 BPSD of 1.72% S fuel oil. There are currently a number of G O-Fining units in commercial operation. Investment and operating costs will vary depending on plant location and crude stock characteristics, but for many typical feedstocks (basis 50,000 BPSD) total investment is about 16.3 million dollars and operating costs average out at 60¢/barrel fuel oil (1989). UOP ’ s gas desulfurization process Another light oil desul-furization process is UOP’s gas oil desulfurization scheme. Unlike the previously discussed G O-Fining process, UOP’s scheme (already commercial) is designed for almost complete (⬃90%) desulfurization of a 630 to 1050; F blend of light and vacuum gas oils (approximate sulfur content of feed—1.5%). Vacuum residuum is neither directly nor indirectly involved anywhere in the process. In almost all other respects, however, UOP’s process parallels G O-Fining. The current plant facility is of 30,000 BPSD capacity with above mentioned feed. Comparison of UOP and G O-Finer costs show that both are of the same order of magnitude and differ markedly only in initial capital investment. This is in part attributable to the fact that a G O-Fining facility requires atmospheric resid-uum fractionation whereas UOP’s does not. Stocks of high-sulfur content are difﬁ cult to crack cata-lytically because all or most of the catalysts now in com-mercial use are poisoned by sulfur compounds. In recent years the trend has been toward processes that remove these sulfur compounds more or less completely. The high sulfur 1100°F Vacuum Bottoms16,600 BPSD4.2 wt % sAB700–1100°F VGO33,400 BPSD2.33 WT % SMIDDLE EAST700°F + RESID50,000 BPSD3.0 WT % S33,100 BPSD0.3 WT % S400°F +DesulfurizedFuel Oil1.72 WT % sTHE GO–FINING PROCESSFIGURE 2C006_002_r03.indd 414C006_002_r03.indd 414 11/18/2005 10:27:10 AM11/18/2005 10:27:10 AM

415contents of petroleum stocks are mainly in the form of thio-phenes and thiophanes and these can be removed only by catalytic decomposition in the presence of hydrogen. The Union Oil Company has developed a cobalt molybdate desulfurization catalyst capable of handling the full range of petroleum stocks encountered in reﬁ ning operations. Even the more refractory sulfur compounds associated with these stocks are removed. This catalyst exhibited excellent abra-sion resistance and heat stability, retaining its activity and strength after calcination in air at temperature as high as 1470⬚F. 8 Cobalt molybdate may be considered a chemical union of cobalt oxide and molybdic oxide, CoO · MoO 3 . The high activity of this compound is due to an actual chemical combination of these oxides with a resultant alteration of the spacing of the various atoms in the crystal lattice. 8 Catalyst life is two to ﬁ ve years. Catalyst poisons consisted of carbon, sulfur nitrogen and polymers. Regeneration is accomplished at 700 to 1200⬚F using air with steam or ﬂ ue gases. The fundamental reactions in desulfurization are as follows: General Reaction C n H m S p ⫹ x H 2 → C n H m ⫹ 2 x ⫺ 2 p ⫹ p H 2 S Desulfurization of ethyl mercaptan C 2 H 5 SH ⫹ H 2 → C 2 H 6 ⫹ H 2 S ∆ H ⫽ ⫺19.56 kg cal/mole Desulfurization of diethyl sulﬁ de (C 2 H 5 ) 2 S ⫹ 2H 2 → 2C 2 H 6 ⫹ H 2 S ∆ H ⫽ ⫺36.54 kg cal/mole Desulfurization of thiophene C 4 H 4 S ⫹ 4H 2 →C 4 H 10 ⫹ H 2 S ∆ H ⫽ ⫺73.26 kg cal/mole Desulfurization of amylene C 5 H 10 ⫹ H 2 →C 5 H 12 ∆ H ⫽ −33.48 kg cal/mole. The change in heat content for all these reactions is negative, indicating that they take place with evolution of heat. The sulfur content in Middle East Gas Oil, a typical feed, is 1.25% by weight. The pilot plant data shows that the heat effect is not serious and whole process can be treated as isothermal. The chemical reaction process on the catalyst is postu-lated to proceed on the surface of the catalyst by interaction of the sulfur-bearing molecules and hydrogen atoms formed through activated absorption of hydrogen molecules. 9 Oil molecules are more strongly absorbed than hydrogen mol-ecules, and therefore may preferentially cover part of the surface, leaving less surface available for dissociation of hydrogen molecules. In the presence of diluent, namely, N 2 , it can also compete for free sites on the surface, and accord-ingly may cause a reduction in the concentration of hydro-gen on the surface, thus giving the lower rate constant when working with H 2 −N 2 mixture. Conversion of the sulfur compounds to hydrogen sulﬁ de and saturated hydrocarbons occurs by cleavage of the sulfur to carbon bonds; essentially no C—C bonds are broken. Residuum Desulfurization The H-Oil-process ( Cities Service ) In order to meet the need for an efﬁ cient method of desulfurizing residual oils with-out the complexities encountered in the myriad of existing ﬁ xed bed catalytic systems, Cities Service developed what is known as the H-Oil system. Although ﬁ xed bed catalytic reactors had been extensively used for desulfurizing distillate oils, desulfurization of residual oil in a ﬁ xed bed reactor presented several difﬁ culties: 1) the high temperature rise through the bed tended to cause hot spots and coking, 2) the presence of solids in the feed and the forma-tion of tar-like coke deposits on the catalyst tended to cause a gradual build-up of pressure drop over they catalyst bed and 3) because of the relatively rapid deactivation of the catalysts, system shut down for catalyst replacement occurred often, on the order of six times yearly. To overcome these problems an ebbulated bed reactor was designed. Figure 3 is a simpliﬁ ed drawing of reactor workings. The feed oil is mixed with the recycle and makeup hydrogen gas and enters the bottom of the reactor. It passes up through the distributor plate which distributes the oil and gas evenly across the reactor. The reaction zone consists of a liquid phase with gas bubbling through and with the catalyst particles suspended in the liquid, and in random motion. It is a back-mixed, iso-thermal reactor, with a temperature gradient between any two points in the reactor no greater than 5⬚F. Due to the catalyst suspension in liquid phase, cata-lyst particles do not tend to adhere to one another, causing blockage of ﬂ ow. Any solids present in the feed pass directly through the reactor. Reactor pressure drop is constant. One of the more important aspects of the ebbulated bed reactor system is that periodic shutdowns for catalyst replacement is not necessary. Daily catalyst replacement results in a steady state activity. Table 2 shows examples of H-Oil desulfurization perfor-mance with atmospheric and vacuum residuals. In addition, investment and operating cost data are shown to illustrate the important effect of feed stock characteristics on overall economics. Cases 1–3 describe processing of three atmospheric resid-ual feeds. The Kuwait Residuum treated in case 1 is a high sulfur oil containing relatively low metals content (60 PPM). Therefore, the rate of catalyst deactivation is low and operat-ing conditions are set to minimize hydrocracking and maxi-mize desulfurization. In fact, only 2–3% naphtha and 9–10% middle distillate are produced. The actual chemical hydrogen consumption is fairly close to the estimated needed to remove the sulfur. For many atmospheric residuals which are not too high in metals, this case is typical to give maximum production of low sulfur fuel oil at minimum conversion and hydrogen consumption. C006_002_r03.indd 415C006_002_r03.indd 415 11/18/2005 10:27:10 AM11/18/2005 10:27:10 AM
416
In case 2, although metals content is also low (⬃40 PPM), hydrogen consumption is exceptionally high. This is due to the fact that conversion was not minimized and 7% naphtha and 13% middle-distillate was produced by hydrocracking. Case 3 is characteristic of high metals content (⬃320 PPM) oils from that area. As noted previously, catalyst deactivation increases with metals content. Therefore, catalyst addition rates are higher, resulting in increased operating costs. To compensate for the reduced catalyst activity, higher operating temperatures and/or residence times are used. Cases 4–6 summarize vacuum residua operations. Desulfurization rates for vacuum residua are lower than for atmospheric. The asphaltenes and metallic compounds reside in the vacuum residuum, consequently increasing catalyst deactivation rates and therefore catalyst costs per barrel. In all the cases depicted (4–6) hydrogen consumption, relative FRESH CATALYSTREACTORIIREACTORIFEED OILMAKE-UPHYDROGENRECYCLE HYDROGENTHE H-OIL PROCESSLIQUIDPRODUCTFIGURE 3TABLE 2H-OIL desulfurisation of atmospheric and vacuum residualsType-Feed (A-atmos) (V-vacuum) Case 1A Case 2A Case 3A Case 4V Case 5V Case 6VSource Kuwait W. Texas Venezuela Kuwait W. Texas VenezuelaFeedstock data—— — —— — Sulfur (Wt%) 3.8 2.5 2.2 5.0 2.2 2.9 Vanadium and nickel (PPM) 60 40 320 90 55 690975ºF⫹ —— — —— — Vol% 45 45 52 80 70 75 Sulfur, Wt% 5.3 3.2 2.8 5.3 2.7 3.2Yield, quality (400ºF⫹)—— — —— — Vol% 99.3 96.0 94.2 94.7 92.9 92.9 % S 0.9 0.4 0.9 1.8 0.6 1.2Chemical H2—— — —— — Consumption (SCI-/BBL) For S removal (est.) 290 210 140 340 170 200 Total 490 670 470 660 640 920Economics (Relatives)—— — —— — Capital inv. est. 6.7 7.8 6.9 7.9 8.3 8.9OP cost, (20,000 BPSD UNIT) 33 40 39 44 46 65C006_002_r03.indd 416C006_002_r03.indd 416 11/18/2005 10:27:10 AM11/18/2005 10:27:10 AM

417SEPARATORFLASHDRUMLOW SULFURFUEL OILMID-DISTILLATEGASOLINEHYDROGENREDUCED CRUDE FEEDREACTORGASESRCD ISOMAXFIGURE 4WHOLECRUDE(117, 000 BPSD)TWO STAGEDESALTER(50,000 BPSD)REACTOR650°F+ATMOSPHERIC CRUDEDISTILLATIONHYDROGENFRACTIONATORC40.1% SULFURMID DISTILLATE1% SULFUR FUELOIL(40,000 BPSD)TO SULFURRECOVERY350°F - (30,300 BPSD)350–650°F - (36,400 BPSD)RDS ISOMAXFIGURE 5C006_002_r03.indd 417C006_002_r03.indd 417 11/18/2005 10:27:10 AM11/18/2005 10:27:10 AM
418
WHOLE CRUDE(100,000 BPSD)TWO STAGEDESALTERCDSISOMAXREACTORHYDROGENCDS ISOMAXSYNTHETIC CRUDEFRACTIONATOR 0.1% SULFURMIDDLE DISTILLATE(29,600 BPSD) 1% SULFURFUEL OIL(40,000 BPSD)C5C4FIGURE 6REACTORSFEEDFURNACERECYCLE HYDROGENABSORBERHIGHPRESSURELOWPRESSURESEPARATORSTO GASRECOVERYLIGHT GASOLINEHEAVY NAPHTHALIGHT GAS OIL650°F+BOTTOMS(1% S)HDSH2FIGURE 7C006_002_r03.indd 418C006_002_r03.indd 418 11/18/2005 10:27:10 AM11/18/2005 10:27:10 AM

bundet

419to that needed for desulfurization, is high indicating that high sulfur content of feed precludes setting of operating condi-tions to minimize conversion. In fact, naphtha production ranges from 7–15%, mid-distillates from 15–23%. The Isomax processes A broad spectrum of ﬁ xed bed desulfurization and hydrocracking processes are now in oper-ation throughout the world. They are characterized by their ability to effectively handle a wide range of crude feedstocks. In addition, some of the processes are capable of directly desul-furizing crude oil while others treat only residual stocks. Rather than discuss each process individually, a compar-ative summary of the major ones is presented in Table 3. There are many other processes which in one way or another effect a reduction in the amount of sulfur burned in our homes and businesses. All of them use some type of proprietary catalytic system, each with its own peculiar optimum operating ranges with regard to feed composition and/or reactor conditions. The hydrodesulfurization process is still relatively expensive (in 1989 more than 75¢/BBL) by petroleum pro-cessing standards. The capital investment for large reactors which operate at high pressures and high temperatures, the consumption of hydrogen during the processing and the use of large volumes of catalyst with a relatively short life all contribute to the costs. In addition, processing costs also depend on the feedstock characteristics. But when one considers the awesome annual alternative of 30 million tons of sulfur dioxide being pumped into the atmosphere, the cost seems triﬂ ing indeed. Desulfurization of Natural Gas Approximately 33% of the natural gas in the United States and over 90% of that processed in Canada is treated to remove normally occurring hydrogen sulﬁ de. The recovered sulfur, which now accounts for about 25% of the free world’s production is expected to increase in the future. Current processes may be classiﬁ ed into four major categories: 1) Dry Bed—Catalytic Conversion, 2) Dry Bed—Absorption—Catalytic Conversion, 3) Liquid Media Absorption—Air Oxidation, 4) Liquid Media Absorption—Air Conversion. Dry bed catalytic conversion ( the Modiﬁ ed Claus Process ) The Modiﬁ ed Claus Process is used to remove sulfur from acid gases which have been extracted from a main sour gas stream. The extraction is done with one of the conventional gas treating processes such as amine or hot potassium carbonate. The process may be used to remove sulfur from acid gas streams containing from 15 to 100 mole % H 2 S. The basic schemes use either the once through process or the split stream process. Figure 8 shows ﬂ ow characteristics of the once-through scheme, which in general gives the high-est overall recovery and permits maximum heat recovery at a high temperature level. Split stream processes are generally employed where H 2 S content of the acid gas is relatively low (20–25 mole %) or when it contains relatively large amounts of hydrocarbons (2–5%). Pertinent design criteria for dry bed catalytic conversion plants include the following: 1) Composition of Acid Gas Feed, 2) Combustion of Acid Gas, 3) For a Once-Through Process, Retention Time, of Combustion Gases at Elevated Temperatures, 4) Catalytic Converter Feed Gas Temperature, 5) Optimum Reheat Schemes, 6) Space Velocity in the Converters, 7) Sulfur condensing Temperatures. TABLE 3Process RCD Isomax RDS Isomax CDS ISomac HDSLicensers UOP Chevron Chevron Gulf R & DGeneral feed type Atmospheric Atmospheric Whole crude ResiduumFeed characteristics Name Kuwait Arabian light Arabian light — Sulfur content 3.9 3.1 1.7 5.5Process diagram Figure 4 Figure 5 Figure 6 Figure 7Fuel oil product Quantity (BPSD) 40,000 40,000 40,000 40,000 Sulfur content 1.0 1.0 1.0 2.2Economies (Relative) Investmenta 9.7 24.5 156.7 10.0 Operating costsb51 40–60 40–60 —a Includes only cost for Isomax reactor/distillation and auxiliary equipment.b Includes utilities, labor, supervision, maintenance, taxes, insurance, catalyst, hydrogen, etc.C006_002_r03.indd 419C006_002_r03.indd 419 11/18/2005 10:27:11 AM11/18/2005 10:27:11 AM
420
ACID GASAIRB+RCWHB2nd HOT GAS BYPASS1st HOT GAS BYPASSMODIFIED CLAUS PROCESSTAIL GASLEGENDB+RC - BURNER + REACTION CHAMBERWHB - WASTE HEAT BOILERR - CATALYTIC CONVERTORC - CONDENSERSL - LIQUID SULFURSLSLSLR1R2C1C2FIGURE 8SWEET GASWASTE GASLIQUID SULFURASAIRSOUR GASLEGENDR - ZEOLITE BED ABSORBERSC1 - SULFUR CONDENSERC2 - ARIAL COOLERAS - ACCUMULATOR/SEPARATORB - SULFUR BURNERR1R2C2C1BHAINES PROCESSFIGURE 9C006_002_r03.indd 420C006_002_r03.indd 420 11/18/2005 10:27:11 AM11/18/2005 10:27:11 AM

421 The sulfur from a claus plant may be produced in vari-ous forms, such as liquid, ﬂ aked or prilled. Form of choice is determined by transportation mode and end usage. Dry bed absorption — catalytic conversion ( Haines Process ) In the Haines Process (see Figure 9), sulfur is removed from sour natural gas using zeolites or molecular seives. Hydrogen sulﬁ de is absorbed from the gas on a bed of zeolites until the bed becomes saturated with H 2 S. The bed is then regenerated with hot SO 2 bearing gases, generated from burning a part of the liquid sulfur production. The zeolite catalyzes the reaction of H 2 S form sulfur vapor. The regeneration efﬂ uent gases are cooled and sulfur is condensed. Liquid media absorption — air oxidation Most com-monly in use in Europe for removal and recovery of sulfur from manufactured gases such as coal gas or coke oven gas, several processes in this category are available today. Typically, the process scheme involves absorption of H 2 S in a slightly alkaline solution containing oxygen carri-ers. Regeneration of the solution is by air oxidation. The H 2 S is oxidized to elemental sulfur. Figure 10 shows the typical processing scheme. The air also acts as a ﬂ otation agent for the sulfur which is collected at the regenerated solution surface as a froth. The sulfur sludge is either ﬁ ltered or centrifuged to yield a sulfur cake which may be recovered as a wet paste or dry powder. Liquid media absorption — direct conversion ( Townsend Process ) The Townsend Process is perhaps one of the newest sulfur removal processes on the scene today. Still in forma-tive stages of development, its potential advantages point to it directly challenging the other conventional system con-taining an amine plant, a dehydration plant and a claus sulfur plant. The Townsend plant offers the added advantage of recovering a higher percentage of the sulfur. The Townsend Process uses an aqueous solution of an organic solvent such as triethylene glycol to contact the sour gas and simultaneously sweeten the gas, dehydrate the gas and convert the gas to elemental sulfur. ENVIRONMENTAL CONCERNS Trowbridge 10 states that reductions in lead content of fuel oil are occurring world wide. Unleaded mo gas originally intro-duced in California in 1975 is now available throughout the U.S., where permissible maximum lead levels were estab-lished at 0.1 grams/gal after 1986. Trowbridge describes additional environmental concerns and fuel treatment proce-dures which are summarized below: • ozone formation—reduction of gasoline vapor pressure • carbon monoxide emissions—add oxygenates to blend • benzene emissions—reduce fraction in reformate by extraction for example. DESULFURIZATION OF COAL Although coal buring accounts for only about ¼ of the nation’s energy, approximately ⅔ of all the sulfur dioxide emitted in the United States is traceable to its usage. And since coal supplies far outstrip gas and oil reserves, interest in coal desulfurization is great. Presently, however, there are no processes sufﬁ ciently developed, either technically or commercially, which have any signiﬁ cant impact on the industry. The following is, therefore, SWEET GASSOUR GASLEGENDRICH SOLUTIONLEANSOLUTIONFILTRATEAIRSULFUR FROTHLIQUID MEDIA ABSORPTIONRAFSKA - ABSORBERR - REGENERATORSK - SKIM TANKF - FILTERSULFUR CAKEFIGURE 10C006_002_r03.indd 421C006_002_r03.indd 421 11/18/2005 10:27:11 AM11/18/2005 10:27:11 AM
422
a brief summary of the more pertinent processes currently under evaluation. A typical high-sulfur, bituminous coal (4.9%) sulfur may be characterized by the composition shown in Table 4. The solvent reﬁ ne process Solvent reﬁ ned coal (SRC) is one name among others given to a reconstituted coal which has been dissolved, ﬁ ltered, and separated from its solvent. It is free of water, low in sulfur, very low in ash, and sufﬁ -ciently low in melting point that it can be handled as a ﬂ uid. In the process (see Figure 11), coal is ground to 80% through a 200 mesh screen and slurried in an initial solvent oil. The slurry is pumped to a pressure of 1,000 lb/sq.in. and passed upwards through a heater to bring the slurry to a tem-perature of 45⬚C. The ﬂ ow rate is between a half and one space velocity. As the material leaves the heater, more than 90% of the carbon in the coal is in solution. (Anthracite coal is an exception.) A small amount of hydrogen is introduced into the slurry prior to preheating. The hydrogen prevents repolymerization of the dissolved coal. A portion of the organic sulfur in the coal unites with hydrogen to form hydrogen sulphide, which is later separated as a gas. The unutilized hydrogen is recycled back to the preheater. After the coal has been dissolved, the ash is ﬁ l-tered at a pressure of about 50 lb/sq.in. The pyritic sulfur leaves the process mixed with the other separated mineral matter. The ﬁ ltrate or coal solution, now greatly reduced in ash and sulfur is ﬂ ash evaporated to obtain the necessary recycle solvent. It has been demonstrated that no make up solvent is required. In fact, an excess of solvent oil, released from the coal structure itself during solvation is recoverable, if desired. The remaining hot liquid residue is discharged and cooled to form a hard, brittle solid of solvent reﬁ ned coal. Table 5 shows a comparative analysis of a raw coal (% S—3.27%) and the solvent reﬁ ned process, along with rough economies. Obviously, solvent reﬁ ned coal represents a marked improvement over the raw starting material. Sulfur content is reduced by about 30%, Ash content by 93% and heating value (in BTU/LB) is signiﬁ cantly increased. However, the cost of these improvements produces a fuel whose cost is higher than low sulfur oil. Devolatilization I carbonization to a low sulfur char United Engineers and Company have investigated the use of a ﬂ uidized bed desulfurization process. Figure 12 shows general ﬂ ow scheme for the label scale process. ASH PRODUCTSASH PROCESSINGASH RESIDUEFILTERDISSOLVERPREHEATERFEED PUMPCOALSOLVENTSLURRYTAN KHYDROGENGAS TREATMENTDISTILLATIONLIGHT OILFLASH EVAPORATORSOLVENTREFINED COALSOLVENT REFINE COAL PROCESSFIGURE 11TABLE 4Coal Analysis (on dry basis) Volatile combustible matter 42% Ash 10% Fixed carbon 48% Moisture —Sulfur breakdown (by % —forms available) Sulfate 0.33 Pyritic 2.53 Organic 2.044.9%TABLE 5Raw coal Solvent reﬁ ned coalAsh 6.91% 0.41Carbon 71.31 89.18Sulfur 3.27 0.95Volatile matter 44.00 51.00Heat content (BTU/LB) 13978 15956Cost ($/Ton) 40.00 120.00Cost (c/MMBTU) 156.00 469.00NITROGENAIRSTEAMGAS VENTFLUIDIZED AND DESULFURIZATIONFIGURE 12C006_002_r03.indd 422C006_002_r03.indd 422 11/18/2005 10:27:11 AM11/18/2005 10:27:11 AM

423 Ground coal (1/16⬙) is introduced into the heated, tubu-lar reactor. Gas ﬂ ow is then adjusted to maintain bed ﬂ uidity. Reactor temperatures approach 950⬚F. The char is then removed through the bottom of the reactor. The factors most directly affecting desulfurization are par-ticle size, reactor residence time, reaction temperature and ﬂ u-idizing velocity. Figure 13 summarizes relationships between extent of desulfurization (expressed as the ratio of sulfur in char to that in feed- ∆ S) and above mentioned parameters. Other processes, in various stages of development include causticized ﬂ uidized bed desulfurization, gasiﬁ ca-tion to high and low BTU Gases and extr action of pyritic sulfur from raw coal. The Bureau of Mines, bituminous coal research and others have sought to remove sulfur (pyritic) by washing, using various techniques including centrifugation, ﬂ otation and magnetic separation methods. None of these has the potential to remove more than half the sulfur and each leads to signiﬁ cant product losses. Bartok et al. 11 point out the advantages of systems which combine advanced processing techniques with advanced combustion recovery cycles. Gasiﬁ cation combined cycle, for example, couples the precombustion clean-up of coal (via gasiﬁ cation) with gas and steam turbines for power generation. As the efﬁ ciency of such plants improves their economics will improve relative to conventional coal ﬁ red plants while having a superior environmental impact (much lower nitrogen oxides, for example). COAL ASH REMOVAL Detroit Edison Company demonstrated the beneﬁ ts of oper-ating with low ash coals. In addition to the lower ash removal requirements, they lead to reduced transportation costs, higher heating values and improved boiler performance. One possible disadvantage of water cleaning method is additional drying which might be required prior to combustion. Ash removal is considered primarily for coal production. Ash contents of oils are relatively small and of natural gas are negligible. The popular separation techniques usually depend on speciﬁ c gravity differences (physical) or froth ﬂ otation (chemical) methods. The dispersing medium for speciﬁ c gravity separation may be water, air or suspended matter in water. Coal speciﬁ c gravities may range from 1.1 to 1.8, whereas impurities typically have speciﬁ c gravities above 2.0 for carbonates and silicates and as high as about 5.0 for pyrites. Bowling et al. 12 state that the levels of ash-forming min-eral matter in most coals can be reduced by a combination of physical and chemical methods, to yield ultraclean coals with ash yields of 0.1–1%. BLENDING WITH PETROLEUM COKE To satisfy a worldwide power production growth rate of more than 2.5% a year, with even higher rates in the developing 10 20 30 40 50 600.60.81.0 1.24006008001000 12000.500.701.00TEMPERATURE (OF)FLUIDIZING VELOCITY (FT/SEC)TIME (MINUTES)FIGURE 13C006_002_r03.indd 423C006_002_r03.indd 423 11/18/2005 10:27:12 AM11/18/2005 10:27:12 AM
424
nations in Asia and Latin America, nations are reviewing their fuel alternatives. One such fuel is petroleum coke (PC), being produced at a worldwide rate of over 50 tons per year in 2000. PC is a byproduct of oil reﬁ ning that can be burned along with other fossil fuels. The cement industry is currently one of the widest consumers of blended PC. When co-ﬁ ring liquid or gaseous fuel, PC will usually require additional control equip-ment for particulate and sulfur oxides reduction. In boilers coﬁ ring coal particulate controls may be adequate since PC ash contents are low, however, the ﬂ ue gas may require scrubbers to reduce sulfur oxides, since PC sulfur contents are high. See table for typical PC properties. The low ash content and low grindability reduce solids handling costs. The high vanadium content may, however, contribute to secondary plumes prob-lems in the absence of sulfur removal equipment. REFERENCES 1. Myerboff, A.A., American Scientist 69, 624 (1981). 2. Lynch, M.C., Chem. Eng. Prog. 84, 20–25 (Mar. 1988). 3. Balzhiser, R.E., Chem. Eng. 74–96 (Jan. 11, 1982). 3a. Hansen, Teresa, Electric Power and Light, November 1996, pp. 15–16. 4. Finfer, E.Z., J. Air Poll. Control Asso. 15, 485–488 (1965). 5. Harris, R.D. and R.G. Moses, “Coal Desulfurization” at MECAR Sym-posium, Feb. 24, 1966. 6. Beckberger, L.H., E.H. Burk, J.H. Frankovich, J.O. Siemssen, and J.S. Yoo, EPRI Report #CS-3902, Feb. 1985, Elect. Pwr. Res. Inst., Palo Alto, CA 94304. 7. Parkinson, J.W. and E.R. Torak, EPRI Report #CS-3808, Feb. 1985, Elec. Pwr. Research Inst., Palo Alto, CA 94304. 8. Bartenope, D.P. and E.N. Ziegler, Enc. of Env. Sci. and Eng. 2, 837 (1975), 1st Edition, Gordon and Breach Sci. Pub., New York (London). 9. Hoog, H.J. Inst. Petrol. 36, 738 (1950). 10. Trowbridge, T.F., Chem. Eng. Prog. 84, 26–33 (Mar. 1988). 11. Bartok, W., R.K. Lyon, A.D. McIntyre, L.A. Ruth and R.E. Sommer-lad, Chem. Eng. Prog. 84, 54–71 (Mar. 1988). 12. Bowling, K.M., H.W. Rottendofr and A.B. Waugh, J. Inst. of Energy 155, 179–184 (1988). 13. Genereux, R.P. and Doucette, B., Power, July/August 1996 . EDWARD N. ZIEGLER Polytechnic University Ultimate & Proximate Analyses and Ash Contend of Delayed PC, Typical Weight %, As ReceivedSpecies Avg. RangeCarbon 80 75–86Hydrogen 3.3 3.0–3.6Nitrogen 1.6 1.3–1.9Sulfur 4.5 3.4–5.3Ash 0.27 0.0–0.6Oxygen ⬍0.1 0.0–0.1Moisture 10.6 5.5–15.0HHV, kBtu/ lb 13.5 12.6–14.5Volatile Matter 10 8–6Ash Analysis, ppmVanadium ⬍2000 500–000Nickel 336 250–50Iron 84 50–50FRESH WATER: see
, WATER—FRESHGASEOUS POLLUTANT CONTROL:see VAPOR AND GASEOUS POLLUTANT FUNDAMENTALSC006_002_r03.indd 424C006_002_r03.indd 424 11/18/2005 10:27:13 AM11/18/2005 10:27:13 AM
425G GEOGRAPHIC INFORMATION SYSTEMS The term “Geographic Information System” (GIS) varies as a matter of perspective, and ranges in scope from specific computer software packages to software, hardware and data, to software, hardware, data, and support personnel. The most exhaustive definition is given by Dueker and Kjern as: Geographic Information System—A system of hardware, software, data, people, organizations, and institutional arrangements for collecting, storing, analyzing, and dissemi-nating information about areas of the earth. While this is an all-inclusive definition, the software packages at the heart of a GIS have their roots in the work of two researchers at Ohio State University, Marble and Tomlinson, in the mid to late 1960s. These men coined the phrase “Geographic Information System,” and defined a GIS as having the following four components: 1. Data Input and the ability to process data. 2. Data Storage and Retrieval with the ability to edit. 3. Data Manipulation and Analysis. 4. Data Reporting Systems for the display of tabular and graphic information. Data input, storage/retrieval, and reporting were (and still are) common to two other computer software packages—Computer Assisted Mapping (CAM) and Database Management Systems (DBMS) which separately managed graphical and tabular information. The development of GIS combined these packages to provide what many refer to as “intelligent maps,” which are maps with extended informa-tion. Extended information can include information from areas such as census, tax assessment, natural resources availability/quality, which may be linked to a map, but are managed as part of a separate database. However, the true distinction of a GIS is Marble and Tomlinson’s third component—the ability to analyze spatial information. The analysis capability enables a GIS to automatically evaluate information from several sources as a function of their spa-tial context. CAM systems may provide some of the same information as a GIS through a series of separate maps, but they require manual interpretation. There are six basic analytical questions which many GIS software packages are able to address, partially or in full. These are: 1. Location—“What is it?”—What types of features exist at a certain place, such as “What is the popu-lation of a given census tract?” 2. Condition—“Where is it?”—Finding a site with certain characteristics, such as “Which agricul-tural fields are within 100 meters of a stream?” 3. Trend—“What has changed?”—Evaluation of spatial data as a function of time. 4. Routing—“Which is the best way?”—A variety of problems to determine paths through a network, such as finding the shortest path or optimum flow rates. 5. Pattern—“What is the pattern?”—A function which allows environmental and social planners to account for spatial distribution. Examples include the spread of diseases, population distribution versus urban development, or targeting specific consumer trends. 6. Modeling—“What if?”—Allow model develop-ment and evaluation, including “Monte Carlo” evaluation, where a variety of factors influence a situation, and their relationship is determined by varying one factor while holding the others constant. A GIS is able to perform the above operations because they maintain the topology, or spatial associations, of the elements of their database. For example, an individual look-ing for 617 East Central Avenue on a map is able to see that this address is between 6th and 7th streets, on the south side of the street. Furthermore, if the individual wants to drive C007_001_r03.indd 425C007_001_r03.indd 425 11/18/2005 10:27:32 AM11/18/2005 10:27:32 AM