Shields and barricades for chemical laboratory ... - ACS Publications

---

in the Chemical laboratory Edited by NORMAN V . STEERE, School o f Public Health, University of Minnesota, Minneapolis, Minn., 55455

VII.

... a

feature

Shields and Barricades for Chemical Laboratory

Operations David T . Smith Superintendent, Protection Division E. I. du Pont de Nemours ond Compony, Chombers Works

There may be a chemist without his own fixed ideas on how to provide shielding for

probably will n o t meet universal agreement. However, there seems to be general agreement that i t is important to provide protection to personnel, performing either experimental or routine laboratory work, from the hazsrda of explosion, rupture of containing apparatus and systems from overpressure, implosion hazards due to vacuum, flash ignition of escaping flammable VapOTB, sprays or emissions of toxic or corrosive materials. Our experience shows that in spite of planning and preparation, these types of mishaps can and do occur, znd that people can and have been injured by the results of t h e miahaps. An analysis of the force and characteristics of t,he hazards involved can help us select effective and economic solutions to the guarding problem. Our decision is simplified somewhat by the fact that if we design protection for the most severe exposure, i t will suffice for lesser exposures. On the other band, we do not want to "averdesign" with a resulting economic waste.

Types of Exposures Although trinitrotoluene and other hazardous high energy materials are capable of detonation, most chemicals explode by chemical processes which are subsonic speeds, and are termed "deflsgrations." Because the time during which the destructive energy is liberated in these cases varies so widely, the total equivalent explosion force of various syetems, as compared to a single material such as TNT, is of limited significance. T N T is capable of detonation and hence of developing its full potential instantaneously. Most chemicals, such as ethanobair mixtures, are not. Even acetylene, dinitrotoluene, and nitrocellulose are caused to detonate only under severe conditions of confinement ofpressure. Accordingly, for most deflagrations, the blast wave pressures are functions of the bursting pressure of their containers. The design of protection for chemical processes e m be based on deflagration

A520

/

Chemicol Education

efTects unless there is good reason to believe that a detonation can occur. IWssiles of great energy may be thrown by all t m e s of explosons. They may consist of broken parts of apparatus, or even parts of damaged shields. For complete missile protection: No unshielded lineof-sight path should he allowed between the exposing apparatus and any part of the observers body. No deflected missile path (ricochet) should be allowed where the angle of incidence with the deflecting surface exreeds 45' (zero angle of incidence is a missde path perpendieulm to the deflecting surface). If we provide adequate missile harriers, the only serious air-blast hazards in bench scale work besides failure of the shield or shield material are: The possibility of ear damage. Self-inflicted injury from involuntary or irrational reaction to the blast. There is frequently the hazard of flash fire exposure, although the explosion hazard may be mild. While shielding of mfficient strength and completeness to protect from missile and blast hazards will normzlly be adequate for flame protection, complete enclosure in a ventilated hood is necessary to insure the containment of flames. Injuries from fires in laboratory work have most frequently been due to the direct proximity of the observer's body to the exposing apparatus. By llmitine,. the mass of the samnle. and avvidtrig the utlsl.ddcd line&ht path l w , w r e n the exposiug i 1 p p w ~ I 1I >1T~Ill(..~tiol. of etnirrcrl fl:zmmnb.e wpm* and :In? p:ut ~f the h w r v e r ' s h d v , tlw lmzard ~ l fl.41t.e f horns I.. i lv~ ~ r ~ ~ l *Jlield ~ ~ t i~o ) ll l ~ r c Tl i ~r l r11 ~ cy c t i u l t l l I I e l involved in a fire following an explosion or other mishap, slow burning material may give good protection a t the fint impingement of flame and heat and during the period when the observer retires to a safe Location. Eliminstian of unshielded line-of-sight paths between source of exposure and any part of the observer's body is also essential to provide protection from splash or spray of toxic or corrosive liquids. Ricochet prntection is also needed. Less dramatic but equally important is the protection against drifting of toxic,

David T. Smith. Grodvofed f r o m lllinoi. Institute of Technology-1930, 8.5. in Firt Protection Engineering; Superintendent o Protection Divirion (mfety, fire protectio, and security), Chambers Works, E. I. du Pon de Nemours & Co. Mr. Smith war recentl) choirmon of o Standards Sub-Commiltee I< develop information for Safely and Fire Pro tection Stondordr on the mbiect of shield ma.

corrosive, or flitmmable gases, vapors, and dusts from the source of exposure to the vicinity of the observer's body. A hood with controlledinward flow of air furnished by mechanical ventilation is required. While inward rates of flow of air a t the face of the hood in the order of 60 fpm* are important to maintain as a minimum, it is more important to study the air flow s t the hood opening with a smoke source under various actual operating conditions. Turbulence a t the hood window opening may prove t o be a problem. Flaw of room air out of a window or door, or displacement of air by an air-conditioning system, may change the picture completely as to effective hood ventilation. Placement of materials and equipment in the hood also have an effect-as s. general rule, the further in from the hood face, the better (with a. 6-in. clearance, s ynod minimum distance). As a. second line of defense, personal protective clothing and equipment ahould be employed: Safety spectacles should he worn regardless of shielding and barricades. Side-shield spectacles offer far more protection than nonside shield ones and are preferred. Face shields or hoods msy be specified for protection during periods when shielding is not effective. Permanently flame-retardant-treated

*The Editor would recommend 100fpm.

Safety

...

cotton clothing is available for use where Hash fire or Hame hazards are present. Shirt sleeve should be rolled down and buttoned a t the wrists; shirt fronts should be completely buttoned. This gives protection not only from flame but also from Hying particles, liquids, and dusts. Laboratory aprons, coats, special jackets, etc., are avltilable commercially for specific needs. Gloves (gauntlet) should be used.

Moferiols of Construction While selection af materials of construction must be done jointly with consideration of design, the following general comments are applicable. While observation of apparatus makes the use of transparent shields desirable, metal shields must be provided to protect against severe missile hazards. T h i ~ applies when materials are in metal containers, or heavy missiles of ceramic or glass may develop. Mirrors, limited-area peepholes, and other dev~cesmay be used to facilitate observation in these cases. Transparent shields may be used for reactions in most laboratory glassware beoause the missiles developed will be very light snd thus have relatively low energy. ASTM test method D 256 measures the relative susceptibility of a sample to fracture by shock :IQ the energy suspended

Vol. 41, No. 7, July 1964

/

A521

by a pendulum breaks a sample notched I in a standard manner in one blow. Another method is to drop balls of varying weights from varying heights on she& rigidly supported, and measure the inchpounds required for failure. Comparative values are as follows: Mmt~Id in.

1.

'/,

2.

I/,

3. 4.

I/,

5. 6.

I/,

7.

doublastrength glass laminated glass plate glass wired glass tempered glass methyl methacrylate polycarbonate Tests

ASTM D 256 (fooepounds) 1. 2. 3. 4. 5. 6.

7.

0.4 to 0.5 12.0 to 16.0

Drop Ball (inch-pounds) 25 110 110 112 582

1086

Considering its cost, transparency, high-tensile strength, resistance to bending loads, impact strength, and slow burning rate, methyl methacrylrtte appears to offer an excellent over-a11 combination of characteristics for litbaratow shields u~ to rhr limtr o i its tnrrngtn. l'&w:~rt,c.naieis 111111.11~ilnmgrr,is s1.lf-e~tir>glli~l11ny .%fIcr ignition, but i~ e ~ s i l ?attdrkwl 11y ~~rwlli(. solvents. It currently is more expensive. Other plastics may be evaluated. (Continued on page A598)

Safety

...

Ignition of 40-g balls of nitrocellulose wrapped in paper masking tape a t s distmrc of 6-in. from a. '/,-in. methyl methacrylate panel resulted in no damage. Ignition of 3-g of nitrocellulose in a 4 o z . tightly stoppered glass jar a t s distance of Gin. from a '/*-in. methyl methacrylate panel resulted in no damage. Four grams, similarly placed, resulted in the fracture of a similar panel. Five grams, similarly placed, resulted in the fracture nf a curved '/,-in. methyl mebhaerylate panel. Steel plate has proved to be about four times as effective as methyl methatrylate in shield work:. i.e... use '1s-in. steel late . for '/3-in. methacrylate; '/,-in. steel plate far 1-in. methyl methacrylate, et,c. Ordinary plate or rolled glass should not be considered for explosion shielding. Wire glass is undesirable for use in shields where there is severe blast effect because, if shattered, the wires may add to the missile damage. Tempered safety glass will shatter to small or grain-size partiela which do not have suflieient weight or cutting edges to he s. major missile hazard. However, if scratched ever so slightly, it may fail on the slightest bending or impact strem. For low-energy shielding, laminatedsafety glass offersexcellent protection, is ineomhrrstible, and does not scratrh easily. Portable shields, including curved and weighted models, should be constructed of not less than '/