Document e7ExargMo739kbMynBg0MxY3q

FILE NAME: Pennsylvania (PA) DATE: 1952 DOC#: PA016 DOCUMENT DESCRIPTION: PA Safe Practice Bulletin - Practical Discussion of the Midget Impinger for Dust Sampling - H P Occupationai'DhsasePreventibn MBI fi iMSMlfli I ^ ibrM %:' 0 C T 6 >?>); :; FHTSSRGH, PRACTICAL DISCUSSION OF THE MIDGET IMPINGER Department of Labor and In Wm. H. Cliesnut - Secretary " A R T IC L E S OR COMMODITIES MENTIONED HEREIN ARE MERELY USED AS IL LU S T R A T IV E MATERIAL AND DO NOT CONSTITUTE IN ANY SENSE AN INDORSEMENT OR GUAR ANTEE ON THE PART OF THE COMMONWEALTH OF PENNSYLVAN IA, THE DEPARTMENT OF LABOR AND INDUSTRY, OR THE WORKMEN'S COMPENSATION BUREAU." a y or M eg and h i^ -p A xxlu d ofindud/uj. OF L A B #  V d&r PRACTICAL DISCUSSION OF THE MIDGET IMPINGER FOR DUST SAMPLING Edited by ROBERT L. HOUTZ Advanced Chemist - . v ' ." /> /'- '5 Safety engineers who are confronted with the task of determining -y^ the degree of hazard of a particular dust within their plant,aro often "5?** ! unfaniliar with the scientific field concerning the best equipment to " v . suit their particular purpose. In order to obtain a technical review of the various available equipment for dust work, and the principlos involved, our department has requested J, B, Littlefield to present a paper on a practical discussion of this subject, Mr. Littlefield's research experience with the U S. Bureau of Mines, and tho part he has played in designing dust-collecting equipment, mako his presentation of timely intorcst. In tunncl-construction work now progressing in Pennsylvania, we find tho ncwly-dcsigned, hand-operatod midget impingcr outfit extromoly valuable in dust determinations of drilling operations, chiefly because of its lightness and portability However, a description of this equipment is being presented , by Mr* Littlcfiold, for the benefit of safety engineers who havo fousd fcMr equipment either too cumbersome, or impractical due to plant operations that will not permit the use of lengthy electrical wiring, or air-sampling hose connections, Mr . Littlefield aslo presents a review of dust sampling practice both in this country and abroad. - lm "PRACTICAL DISCUSSION OF THE MIDGET IMPINGER FOR DUST SAMPLING John B, Littlefield THE MINE SAFETY APPLIANCES CO. PITTSBURGH, PA, "Boiled down to fundamentals, the question at hand is* 'What measure of air dustiness will show whether a particular atmosphere has a safe or dangerous concentration of dust? The particles w h i c h are dispersed into the atmosphere by industrial processes are very small and those that remain in suspension for any time are extremely small indeed. The latter usually range in sizo from ten microns down with b y far the greatest proportion one micron or less in size. In selecting a mode of attack on this problem it is necessary to consider the properties and characteristics of the material with which we are dealing. A thousand particles, one-micron in size, of a given substance have the same mass as one-ten micron particle# A one milli meter sphere of quartz, weighing approximately one and a third milli grams, if crushed into one-micron particles would result in ono thousand million particles and this operation would increase the surface area of the nntcri&l by one thousand times* One more point to remember is that the larger and heavier a particlo is the more rapidly it falls and is, so to speak, removed from circulation, A ten micron particlo of silica, in still air, falls about 70 fc-ot an hour and a onc-nicron particle will fall only about ten inches in the sane length of time. The selection of the sampling method is greatly influenced by the method,'used for the determination or estimation of the concentration or relative concentration. Four possible moans of measuring the relative amount of dust are obviousj first by mass or weight, second by chemical analysis, third by measurement of surface area and fourth, by number  2 v-. r-i; concentration. Determination by weight is not desirable for several reasons* First, because it is too greatly influenced by a small number of large particles and affected very little by a large number of small particles. This method is applicable to the determination of poisonous dusts which are quite soluble or to dispersions of partioles of uniform size such as fumes. At the present, however, we are mainly concerned with fibrosis-producing dusts such as silica or quartz and it is,well known that a given weight of very small particles is muoh more harmful than the same weight of larger particles. It must also bi> remembered that particles 10 microns or larger in sizo rarely entor tho lungs, but aro caught in the nose and are either discharged or swallowed, Anothor objection to determinations by weight is that ordinarily very largo volumes must bo sampled in order to obtain sufficient material to weigh. Chonical analysis introduces tho some difficulties as deter* mination by weight in that large particles influence the results unduly, - Available methods for quartz arc slow and none too satisfactory. This meansK however, is eminently satisfactory for determining the toxic netal dusts and fumes, \ Determination of surface area would probably yield the most desirable results but unfortunately no practical and convenient methods aro available. The surface area of the quartz dust inhaled is probably the greatest factor in its fibrosis producing effect* Thus, the choice of a suitable measurement of air dustiness has narrowed down to terms cf number concentration, at loast for fibrosis- producing and nixed dusts. For such a measurement,means must bo found for counting the particles in a given volume of air. The smallest particle readily visible to the naked eye is about 100 microns in diameter, so a microscope must bo usod to render tho particles visible, V 3 Microscopic Counting, however, is not the complete answer to the que s t i o n The m i n difficulty is that with higher and higher magnification more and more particles are made visible; this number is greatly increased with darkfield illumination in which the sample is illuminated from the side and the tiny spots of light reflected from the partioles are observed instead of the particles themselves. Darkfield illumination is best illustrated by a simple example; we do not see the dust in this room, but if we darken the room and look at a beam of sunlight entering through a small opening we say we see the dust in that beam of light* What we really see, is the light reflected from the particles of -dust* most of which are much smaller than the 100 micron particle which is visible to the naked eye. Working with very high magnification is troublesome and counting the small particles found is extremely tedious* Slight changes in technique introduce wide variations in results and no matter hew refined the technique one can probably never reach the point where every particle may be counted. For very high magnification, the sample must be deposited dry and unless very large areas are to be counted, the deposit must be uniform. Consideration of these difficulties leads to the practical solution of the problem. Is it necessary to collect all the particles in a given sample or to count all of then? Is it even necessary to count all that nay be found in a given small portion of a sample? Instead \^f attempting the impossible or impractical, an empirical method is used*. Instead of counting all of the particles, one counts the portion that nay be seen under prescribed conditions. The empirical results thus obtained under the selected conditions must be reproducible; they must be cor related with studies on the hygienic effect of the dust in question. Once the method has been established as a measure of various conditions. - 4 and when sufficient data to show the effect of these conditions on the health of workers has been accumulated, we are prepared to study.air dustiness and to say whether a given condition is relatively safe or harmful* llcw that an empirical method of determination of relative number concentration has been selected as most practical, a method for collecting the sample must be devised* The fact that dusts are dispersions of solid particles, varying greatly in size and composition, renders the collection of a sample extremely diffioult* Gases are readily collected by bubbling through a suitable sampling liquid, but with a solid, tho liquid must be absorbed upon the solid surface before wetting can occur* Only a few per cent of a fine dust can be removed from suspension in air by bubbling through a liquid* Uany ideas have been put forward for collecting dust samples from air or for determining the amount of dust in suspensions with out collection, but most of them havo been dropped for one roason or another* To better understand the problem, it will help to discuss a few of these methods very briefly, and point out some of their disadvantages. Perhaps the simplest method of studying dust conditions is by observation of the amount deposited as on rafters or beams over a given period, or by simple observation of tho dust in air as illuminated by a beam of light. This gives a rough idea of whether the location is dusty or not but practically no quantitative information* ^ The next step, the sedimentation sampler, is based 111)0 11 the same idea* A container with removable ends is opened at the sampling location and flushed out by moving through the air. It is then closed and allowed to stand in a place free of drafts or temperature changes until all the particles, or at least all that can be seen with a nicrosoope are settled. Then a slide which had been placed upon the - 5- bottom of tho coll is removed and the number of particles per unit area aro counted with a suitablo ni^roscopc, This method has the disadvantage of requiring a long settling timo betwoen samples -which may bo collected with each apparatus# Tho necessity of placing tho apparatus quiokly in tho position for sottling in a looation freo of drafts and tonporaturo changes renders it impractical for field uso and tho f&ot that it is a "grab" sampler makos single determinations of little valu unless tho dust suspension is known to bo uniform The next principle of intcrost is that of filtration* The air to be sampled is drann through cithor an insoluble or a soluble porous % medium such as a paper thimble of a layer of granular material suoh as sugar, anthracene or salaeylic acid* Samples collected on insolublo filters could only be determined gravimotrically with the attendant errors due to particle size variation inherent in suoh a determination* Tho samples collected on soluble filters could be counted after solution of the filter material# Some crystals used for filters contained much and variable amounts of insoluble impurities# Theso filters are difficult to preparo and inconvenient to handle and the increase in resistance as the pores become clogged renders accurate measurement of sample voltio difficult# Although washing is a vory inefficient method of removing dust from air, many dcvioca working on this principle have been tried i. One of the best known is "The Palncr Apparatus") that uses a spray for in timately mixing air and liquid# Such devices,in addition to being in efficient are very cumbersome, fragile, and difficult to clean# A moro recent development is the Thermal Prcoipitator# which operates on a unique principle# The sample is caused to pass betwoen two parallel covcr-glasecs; between and parallel with these glasses is;% heated platinum wire which is perpendicular to the direction of flow of the sample, This device takes advantage of tho phenomena that a body warmer than the gas about it is surrounded by a dust free zono This zone is intercepted by the surfaces of the cover-glasses which aro kept cool by masses of notal in close contact. Also since particulato matter in warm gases tends to deposit upon oool surfaces we have two forces which act to deposit the dust from the sample upon tho two discs of glass The sample appears os a narrow strip on each glass parallel with the wire These samples may be used for count or particle size determination after they are suitably mounted upon a slide, Tho Thermal Precipitator is used * in England and South Africa for oheoking other methods and to some extont for routine sampling. It is high in collection efficiency excopt possibly for the larger particles. Its main disadvantages are the difficulty in preparing and mounting the cover glasses and the snail sampling rate possible for this instrument, about 6 ml per minute, Tho most recent development, the Electrostatic Precipitator, which will be described in detail later by Ir. Barnes, is without question the most satisfactory device for complete separation of solid dispersions from gases. It collects a large sample in a short time and if properly constructed and operated has on efficiency of 100 per cent* ,7c come now to the most widely used class of dust collecting instruments, the devices which operate upon the impingement principle. Three types of inpingers arc fairly well lmown; The "Koninctcr" ^ the "Cwcns-jct dust sampler, and tho "Greenburg-Snith Impingcr," The Kamineter was developed in South Africa about 1916 and is widely used both there and in Canada today. It consists of a snail pump with a spring operated piston which './hen released draws a zb or 5 cc senolo of air through a small circular orifice that faces and is close to a glass plate which is prepared with an adhesive film# The dust in the air sample is impinged upon the adhesive film and sticks there. - 7- The Owens-jet Sampler is very similar in principle except., that it uses a clean class sample plate and the dust laden air is humidifiod in a chamber, then upon being drawn through a slit orifice the reduction in pressuro causes condensation of noisturo on the particles, thus causing thorn to adhero to tho glass surface. The piston in the pump is actuated by hand instead of by a spring. Both of these instruments, cspooially the Konimctcr have many desirable characteristics. Thoy also.have several important shortcomingst both are spot or grab samplers, and unloss a great many samples arc collected at ono location, a fair, averago of tho concentration of dust will not be obtained. The samples oro obscured by deposits of noisture or oil droplot6 and the presence of smoko in the sample either makes the sample spot hazy or results in unduly high counts if darkficld illumination is used. There is also a possibility that with the Owcnsjet dust sampler some of tho sample deposited at the beginning of the piston stroke is subsequently blown off* At the Bureau of lanes in 1922, during a joint investigation of these and other sampling devices by tho U # S, Public Health Servioe and other interested parties, the Grccnburg-Snith Inpingcr was developed. Recent models of this inpingcr consist of a graduated tubular glass flask, 5 0 m or 2 inches in diameter and 210 mm or inches high!, a number 10 rubber stopper with one hole accommodating a combined suction connection and inlet tube. The inlet tube extends to within 5 mm of the flat bottom of the flask where it is drawn down to a 2.3 millimeter nozzle. In operation, the graduated flask is filled with about 75 co of water or alcohol and the side a m of the inpingcr is connected with a source of suction providing a flow of 1 cu ft per minute at a head of about 39 inches of water. Suction is provided either by a portable electric pump or by an air ejector. Both are cumbersome and heavy and of - 8- course limited to use only where olcctric power or compressed air is available The nocossary oquipment with hoses, wrenches, pipe fitting or motor, pumps and wire usually exoceds fifty pounds Halting connections interferes with work being done in the vioinity v:here it is desired to collect the sample Despite theso difficulties the us? of the GreenburgSnith Inpinger has grown and for some yoars has been considered in this country the standard method for collecting sanplos for measurement of air dustiness* It has achicvod this success because in conjunction with the standard light-fiold technique for counting, a strictly empirical method,, uniform and roproducablo results could be obtained under similar conditions with different impingers operated by different persons and with the samples oounted by independent technicians It is true that this enpirioal method has boon subjected to some variations such as the introduction of different types of colls and the uso of micro-projection nethods of counting, but these changes wore adopted only aftor long study showed that the results obtained with a variety of samples checked results obtained by the previously accepted tochniquo Although the advantages of the inpinger method greatly out- \ weighed its disadvantages certain changes appeared imperative if the method were to continue the standard Changes in our ideas of safety and suitable working conditions and in our compensation lavs awakened widespread interest in dust hazards with the result that many now workers entered the field The Federal Bureau of lines has long led the way in the pro 1 motion of health and safety in the mining industry and about 1930 this  9- the need of a compact,, lightweight sampling device which could bo used at any place independent of any power supply, A hand operated inpingcr was tho obvious answer but the amount of energy required to create a flow of 1 cubic foot per minute against a hoad of 39 inches of water was considerable more than a man could bo asked to expend except over very short periods# This was tried but tho change did not reduce oither the weight or the size of tho apparatus# You,- no doubt, are wondering what is sacred about a 1 oubio foot per minute sampling rate especially since the sanplo is. ordinarily diluted to several hundred cubic oontimeters, thon only three or four cubio centimeters are used to fill several counting cells and finally the dust in only 2% cubio milineters of the liquid or about one tonthousandth of the original sample is counted* That thought proved to be the answer to tho questionj a continuous sample over a suitable length of time is most desirable but the rate con be reduced greatly provided it is known and is oonstant* Then if the size of tho impinger, the volume of collecting fluid and the final dilutions are all reduced proportionately one can then oount a larger proportion of the sample# A sampling rato of one tenth of a cubic foot per ninute was selected end several experimental "midget inpingers" were constructed in 1936 and 1937# Experimental work showed that an inpinger with a i m nozzle operated at a head of 12 inches of water would give a sampling rato of 0,1 cubic feet per ninute and. would check tho 1 cubio foot per minute impinger as woll as tiro of tho latter check each other# Two difficulties were encountered in construction of the midget inpin-crj first small capaoity pumps are hard to find and are very low in efficiency and second, it is hard to maintain a constant head and consequently a constant flair by regulating the crank speed. The first difficult:/ was not by constructing a four cylinder radial pump so that I 10 suction was always providod by one oylindor while tho crank was rotated* Tho second problem solvod later when the ilino Safety Appliances Company undertook to make tho midget impinger available to industry* A vaouun regulator was dovised as an integral part of the pump and thus a constant head is provided at any crank speed above fifty to sixty revolutions per minute. This arrangement works so satisfactorily that the oronk m y bo Jerked or even reversed during operation without changing the flew evon momentarily* The entiro apparatus as finally developod, complete and ready for sampling weighs less than nine pounds and occupios loss than one-third cubic foot of spaco* Finally, 1 want to point out some of the precautions that Bhould be observed in using the L'idgot Impinger* Beoause of the small sample volumo noro care must bo oxercisod to prevent contamination* Tho flask and nozzle should be washed, rinsed and assombled, then rinsed several tines through tho intake tube and side a m finally rinsing with the dust free collecting fluidj then the cap and side arm stopper which haVo been washed and rinsed are put in placo* The impinger should be dried and the joint between tho stoppe-r and tho top of the flask protected with a turn of friction tape or scotch tapo* This is important bocauso w h e n vtho sample is returned to tho laboratory for counting, the out6ido can bo washed and dried, then when the tape is removed the joint is clean and dirt will not drop into the sample when the otoppor is removed for dilution* The end of tho hos that connects to the side a m of tho flask should bo rinsed after each days use, and the cap and side arm stopper should be put in a clean metal case which is carried in tho pockot when they are removed to collect a sample* A water-tight hunter*s match case ;S3e is ideal* 1 Contamination is avoided by diluting the sample in the original H - 11- collection flask and by removing portions for counting with a olean pipette. Of course, if the concentration of dust is very high, a portion must be removed and diluted in a volumetric flask* Alcohol, either 95 per cent ethyl or a practical grade of normal propyl is reoomnended for use as the collecting medium in either the large or midget impinger. The advantages of alcohol arei(l) greater ease of cleaning the apparatus such as impingers and counting cells, (2) Tfetting of some dusts suoh as coal,and better dispersion of aggregatesi (35 Reduced solubility of most mineral dustsj (4) Freedom from danger of freezing samples in winter, and(6) Solubility of the o i L m i s t eollocted by impingers near rock drilling operations where oilers are used in the oompressed air lines for rock drills* Tftien using alcohol a set tling time of 25 minutes for the sample in a Sedgv/ick-Rafter counting cell gives results which agree with those obtained with a 20 minute settling time for water. It is important to avoid using mixtures of alcohol and water or denatured alcohol in impingers &e sono of tho sample may be carried out of the flask by the foam which theso mixtures generate*