Fire-proofing wood with chemicals

j I J, h1


rII-Imp1OOrflNG WOID WITH CIItMICALS
1933















S..--------


No. 11145






UNITED STATES DEPARTMENT OF AGRICULTURE
FOREST SERVICE
FOREST PRODUCTS LABORATORY
Madison, Wisconsin
hI Ceoperamion with the Univrerilty o Wisconsin







FIRE PROOFING '"OCD "ITH COIMICALS*


By
0. M. HUNT
In Charge, Section of Wood Preservation
and
T. R. TRUAX
Senior Wood Tec'hnologist



T'.. nc-d for increasing the resistance to fire of lumber and
structural tii,'ber is obvious. 'thile it is true that the fire hazard of
wooden structure. of widely different typos can be reduced by proper
methods of design and construction to the point where they are reasonably
safe, it is also true that effective and economical treatments of wood to
reduce its iLiflan.-iablllty would increase the fire safety of wood struc-
tureq :',nd p>- rmit the use of wood in places where its combustibility now
excludes it*

2ho problem has been recognized for centuries and hundreds of
different fire-retarding coatings and impregnating solutions have been
proposed vit'.in the last 200 years& The majority of the proposals have
been based u-.on adequateae knowledge, and are ineffective, too expensive,
or for GoiaC- othi.r rea-on impractical. Some of thu impregnation methods.
however, h',ve stuficient effectiveness and practicpbillty to have found
commercial 'u-e. It is possible now to buy commercially fire-proofed wood
so resist nt that it will not of itself support combustion. Effectively
treated *,oo1 cai be destroyed only by continuous exposure to heat from
an outside source; when the heating is discontinued the burning ceases.
In crntra.t with untreatedd wood the difference is of the greatest prac-4
tical significance.


Fire-Proofed Wood Defined


T'-c term "fire-proof" as applied to structural materials is
technically ijicorr-ct. Even stone, concrete, and metals arc destroyed.
when ex--osee, continuously to sufficiently high temperatures. Wood treated
by the moit :ffictive methods known is destroyed in time by chemical de-
compos.ition if the .urroun.dinj temper tures are kept sufficiently high.
Popul-r u-s,:, hor.,-vcr, ploys the tcrms "fire-proof", "fire-proofed",
and'If ir c-..I-0oof 1 in connection with wood in the sense that the treated
wood will not u.mu,,ort flame or contribute to its own destruction. Con-
venient -iord, th:t are technically correct are lacking, especially as
substitutes for "fire-proofed" and "fire:-proofing." The popular terms wills
thc-riforc, b" u*scd in this article with the meaning defined above.

*Pa Iib.-. .::. ip Proc. of the ith Pacific Science Corgress. 1933.
R114 1-







Comaerciail Imjregntion Trcatnr-nts


Although o good quality, commercially firc-proofed wood, treated
by imprc-:n.';ion -rocesses is -,vailcble, it is not extensively employ. 1.
Its use in iTorth Amnerice is limited for the most pert to wooden floors,
doors, anc' trim in lalp- office buildings of fire-proof construction in
the city of 1Te,, Yc-rk., Its limited use is due not to lack of fire
re!-ist.'-nc-., jut to its relatively high cost. A contribute..; limitation
is the fact that fire-proofing processes are not standardii::L1. and gnerally
kro'wn. r2o various fire-proofing crmuanies presumably use different
formula nlthouh it is not possible to be cc-rt-.in of this for the form,'i"-
are kept secret. There is also consider ble secrecy as to their meth.od.s
of imTprcLn--,tin- the wood and of drying it after-.ards. A further limits:;
factor is the 1-ck of eccerted standard methods of m.-.'ring the effective-
ness of the treo-'.nment. The obvious advantages that w!ouldl ,ccrue fro:;- the
widesi-re- -,o of adequately fire-proofed woo. justif- considerable f-fort
to overcome .thee obstacles,


Research in Fire-Proofing


Reseearch has not been entirely lacking, in the field of fire-proof-
ing woo:d, -- a review of the literature on the subject till showo.-,'3,
The- various invosti:ptions reported have contributed much to the gener!J.
kno,.lerl;e of !.' ubjOct and have laid foundations for further progress,
but none have resulted in a :;fficiently; convenient satisfactory and inex-
pen1ive method of fire-proofing wood, nor have ,,ny p.rouced testing and
control nmeth'iods that are .-i, rlly accepted .nl. satisf-, the needs of
prospective pur.chasers of fire-proofed wood. Unless the b nefits of a
more extc:-.sivc use of fire-proofed wood -reo to be abandoned much ,.ii-
tional research must be done,

"'.,. U. S. Foroet ProAucts Laboratory at Madison, !is. early
recognizeoC the importance of" fire-proofini studies ;*-.L undertook some work
in this field. -:L results of its f.i.di..-- in these earlier stuliec were
published, in 1915.-2 The continual dcmsnd for information on the subj.ct

IGrratt, G. A. Amer. Lbrman., To.2713, pp.5'-7; :To.27l', 7n.52-3, 19'7.
-2Ingberg, S. HE. S-fety Vn., V.53, lTol, pp. 29-35, 1927.
-Truax, T. .. :vd. Harrison, C. A. Trpns. LAT r. Soc. I,-c'-. En-rs.
N'ood In&., V.52, No7I, :.p33-40, 1930.
14
--toolson, I. r. Rep. Proc., Inter. Fire Prey. Congr., p.257, 1904.
-Prince, R. Z. Procec.ings of the 1T1tl. Fire Protect. Assn., pp.lOS-
5S, 1915.



R115 -2-



J







:.,ni th cr.,ieu". nc:-' for imprcvemcnt in fire-proofing and fire-testing
tLchnic c u;;.' Lc Laboratory to take up the investigation again in 1927,
':r For.:.t Po.-'uicts Labor:tory cnt. rt:A.in no dulu'ion that its studies
i'tll e '."' thu ob!tncles to th.e .xtinsive use of fire-proofud woos,
but hor.i.' t...:t its c,:itributionc, added to those of other investigators,
villa bri.- te desirerl objective definitely nearer. Several reports3,'2'l
of th,. pro Trcss made in the recent invcstlg-tionq have thus far been
7uli: h. T:'.! 1"orl' -'ill undoubtedly continue for several ye.irs,


Fire-Rct rdPnt Chemicals


Alt iou-ho fire-proofing comp:.inies do not make public their fcrmula.s,
there is rca c,:-. to believe that thE monobasic and dibasic phosphr'tc. of
:imnonia arc quite .cncarally use5 in impre..gnation troccses throughout h.c
world :-r.n.I ts.-.,t on- or the othc:r constitutes an important part of most c,--i
con'.crcial formulas used in fire-proofinr byr impregnation. Other ch:mic?.ls
are rdcl'. for d]1f.."rcnt purposes according to the ideas of the respective
companies or "nvcntors and there m:y be some good formulas in usc that
contain s1o a.monium phosphates.

T"-L. chief advantage of the ammonium phosphates is th'ir hic;h
eff.ctivc:5cs.; in retarding the combustion of wool., coupled '-ith the fact
that they do not rive the wood undesirable properties to a sufficient
dc-grL to make tct use of the treated wood impracticable. There -re many
chmrricUl- tl-it have sufficient fire-rctarding ability to be used,_ if it
were n-t for their undesirable properties, Calcium chloride, for example,
attracts moisture e*nd under conditions of high humidity '..ould tnd to kcc.p
the wccd too d mir if used in sufficient quantities to gct gooA fire
resistance. So-ec chemicals attack the wood and reduce its strength, some
destroy or int,.rfere with paint ani varnish adhesion, som,. discolor the
wood or r,.a.: it too hard to work with edged tool-,, some interfere with
gluir.n, or -r, -_-,isonous or too expensive, and practically all that arc
in ccm...:-rcial use are too easily leachLd from thewood to remain permanently
effectivee "'-n '.Md out of doors. Altogether, the requirements arc so
ri-id that no sin-le chemical thus far meets them. Mixtures of chemicals
offer -,rcat-.-r ;'romir.e of meeting these requirements than single chemicals
;r.4 it is quitc possible e that, in due course, highly effective, nonleaching,
r.Jd other,,ise acccv.tablo treatments will be possible at moderate cost.

In m-n.-y places where fire-resistant wood is desired, resistance
to iec-, -nra ic-.sr ct, is also important. In developing fire-proofing
formuln,,. thi. T hr.uld be taken into consideration. It should not be
"iffic'-lt to include in the mixture 0ome tcxic chemical which by adding

r "-'u-.t, "-r-i-on, acchlr. Proc. Amcr. T'ood-Prcs. Assn., V .26, '.
l O -, 9ir; e.27. pr.Oi-gl, 1931; V.2 P p.71-93, 1932. r.29, 193t,
Tru-,'", -. -'. fire-proofine- of w-rod, Prrc. 1,','tl. Fire Pr-tect, A..-,
pp. IS?- ,L, 19"1.






resistance to dec,' and insects would m.tcrirlly incr r-ase the value of
the treated "oo(. and thus help justify the cost of treatment,

One of the chief disadvantages of the more desirable of the
present firc-retjrrdlng chcmicels is their cost. Th.- 'uzntity required -,-er
cubic foot of 'rooJ is so large that the cost of the chemical has probrbb!;
been the lar-cst single item of cost in the production of co.-m. rci:lly
fire-proofed vooc. For a high degree of effectiveness approximately
four pound" of the nmoro e effective chemicals are required per cubic foot
of wood, or orrh.-ps 300 pounds per thousand feet board measure. T'... count
of the chrac.c.l [lone rimy, therefore, increase the cost of the lumber from
$15 to $25 'cr thousand feet. When the necessary treating, drying, -nd
handling costs, 1lant depreciation, and profit are added, the selling:
price of thorouWhly firc--roofed wood may, even under present conditions,
easily exceed that of similar untreated wood by $40 to $50 per thousand
feet board measure. The cost of the treated wood then becomes so high
that it shearply limits the extent to which fire-proofed wood is used.


Application of Fire-Proofirng Chrnic-tls


I methods of applying fire-proofing chemicals to rood vary from
simple 'ur'ff,.ce applications and coatings to thorough impregn-tion,
Obviously there can be a wide r-.nge in the costs of such treatments,
the degree of effectiveness obtained, the ease of application, and the
equipment required. At the one extreme is wood that ignites and burns
almost as can'ily as though untreated end at the other extreme wood that
ignites at higher temperatures and burns only when a considerable quantity
of he-t is supplied from an outside source. There are intermediate
treatml-nts rhich reduce the hazard of untreated wood and which may
ultimately find a large field of usefulness. The choice of a method
of protocti:-L.i: wood against fire must be based upon the requirements of
the use and such .rrctical considerations as cost and convenience.

Surface applications vary considerably in effectiveness, but
are not dependable where severe fire exposure continues for some time. A
thick coatin.; of suitable composition may delay ignition and spread of
flames for a long time when the fire source is small, but when exposed to
a large fire their retardingeffectis only temporary, On the other hane.,
coatings can be anplied to "ood in place, or to fabricated products, by
brush, s-:r.a-, or dipping methods with very little equipment. They, arce
relativel- chc.'i- and their use requires no technical traininri or expericncc.
Coatinys off.'r the opportunity of reducing the hazard of wood by dccrcas.. '".:
the n-imbor of fire"-; starting from small sources and by decreasing the rate
of sprca'. in the early stnge- of a fire.

The methods and apparatus used for injectin.r: fire-proofing
chemicals into .:ood arc practically the same as for injecting preservatives.
The wood is sealed within a treating cylinder and the treating solution
forced in by means of pressure. If the treated wood is to be used for
purposes that require no cutting after treatment, a moderate degree of


R1145


-4-







-enetration may be sufficient even though a core of woor' in the center
of each iiece may remain unimpregnated. When the lumber must be cut ur
Into smaller pieces after treatment, or must be moulded or shaped in such
a way that a considerable depth of wood is removed from the surface,
complete penetration is a necessity. The effectiveness of the chemical
would be largely dissipated if the subsequent cutting exposed any con-
siderable areas of untreated wood.

The deep-penetration requirement complicates considerably the
problem of the commercial fire-proofer, for wood is generally much more
resistant to deep impregnation than is commonly supposed. The sapwood
:f most commercial woods is not difficult to treat and when all-sapwood
lumber is available it offers no serious treating problem. Most lumber,
however, contains some boards that are all heartwood and many boards of
mixed sapwood and heartwood. Such material, especially in species with
highly resistant heartwood like white oak or red gum, thoroughly tests
the skill of the most experienced treating engineer. Complete impregna-
tion of the heartwood in lumber over one inch in thickness requires
great care and watchfulness aa well as skill. The temperatures and
pressures used may easily be made do severe that the lumber is ruined
during the treating process, These strict limitations naturally affect
the treating costs,


Evaluating Fire-proof Treatments


Two of the difficulties in the way of more extensive use of
fire-proofing treatments for wood are the lack of standard methods of
test and the difficulty of interpreting the results of tests in terms of
pTerformance standards, Until adequate progress is made alonr- these lines
there will be misunderstanding, misrepresentation, anda slow appreciation
of the value of fire-proofing processes.

Tests on the inflammability or fire resistance of wood may be
grouped into two classes: Tests on small pieces or representative samples,
an' tests on built-up assemblies under standard time-temperature conditions.
Tests on built-up assemblies are intended to measure the performance of
m'-terials under approximate fire conditions, while tests on small samples
are useful as measures of one or more of the properties of fire-proofed
w'z.od and, in treating-plant control, as measures of the thoroughness of
treatment'* D, 3'ilt-up assembly tests are expensive and are not adapted to
systematic checking of commercial treatments nor to extensive routine
laboratory or development work.

Alth-o'-ah a large nu-ber of tests have been used or proposed for
s-.all piecess or samples and most of them show differences in fire resis-
ta-.ce between untreated and fire-proofed woo&, the results of many are of
ow.ibtful v.lue because of unr-controlled conditions and factors in the test.
Furthermore, there has frequently been failure to use sufficiently
sirnificar.t unite of measurement and little or no correlation of the


4nr' .






condition- of the test to fire conditions, The fir-tub..: test d.:evelop. c
by the U. S, Forest Products Laboratory,j| has been four! valuable in
studying the firc-rcoarding effectiveness of various chemicals, The tcst
has considerable merit from an inspection and. control standpoint and is
receiving consideration as a standard method by intorcstcd agencies,
The oxtecnt to which it will prove to be an Indicator of performance in
assembly tests remains to be determined.


Problems of the Industry


The major problems facing the firc-proofing industry arc the
improvement of methods of trcatmcntf the lowering of cots; the st.-1d:'rd-
ization of mn,.tzrials, processes, and test ncthods; and the clinin-rtion of
secrecy, If those things can be accomplished, tho industry can grow co-
; sidorably". Xf they cannot, be accomplished, cx-r,[nion is not likely to 'c
groat or rapid.

The rosoarch ,,orker in wood fire-proofing, whether .within the in.
duatry or in independent laboratories, faces several specific problems
whose solution will supply the foundation for the expansion of the induftr',.
Among these -rc: (a) developing nee test methods or improving: old ones;
none of the mot ods yet devised has boon accepted as a standard measure of
fire resist-nico although several Gf them give very useful and indicative
information; (b) finding chemicals, or mixtures of chemicals, that ar,
froee froni other objections and are very cheap or arc so effective that
they can be v.scd successfully in much smaller quantities than those now
available; (c) finding chemicals or mixtures of chemicals that are not
only effective against fire and. moderate in cost, but also effective
against decay Fand insect attack, and that do not loach rapidly in out-
door uses.

biestio-is of more purely scientific interest that a ,pcar of loss
immediate practicability, but may in the long run contribute greatly to
solving the fire-proofing problems arc: Just how does wood burn? That is:
What chcmic-.l end physical reactions are involved? Hour do fire-proofing
chemicals retprd combustion? and, Why are some chemicals so much more
effective than others? All of those questions are exceedingly interesting
to investigate, but years may be consumed in finding their answers.







g
-Truax, T. R. and Harrison, C.. A. Proc. Amer. Soc. Testing Materials,
V.29. Pt. 2, pp.971-gg, 1929.




Lln.iE RH iTY rj I' ORI[A

3 1262 08929 0745

I


II
*
i