XII .
Fartber Observations on the Processfor con7verting cast ; intv rnalleable Iron In a Letterfro Thomas 33ed-does , M. 1)* to Sir Joseph Banks , Bartv P 1R S Read May , 79XX DEAR SIR , bINCE I described to you the process known among the workmen by the term puddling of iror ) I have several times reconsidered the explatlation I attempted in the same lettern of the pllaenomena it presents .
WIy explatlation could not indeed but be in great measure conjectural ; and subsequent reflection excited in my mir2d a very lirrely wish to ascertain ) in a decislve manner , the rlatllre of the process .
The following experiments mrill ) I flatter myself , serve to determine thediegree of confi denceXwith which the pWrincipal points of ny theory may bereceived , though they will not afford a soltltion of all the ( uestiOl/ lS which my former communication might suggest to an acute philoso-pher .
Tizey were urlciertaken in order tc : &gt; ascertaDin , .
whether any elastic fluids are really extricated during the conversion of cast : into malleable iron ; and 24 what is their nature ; and g. whether they vary at diSerent periods of the process , as I con cluded from the appearances in the furnace .
It seemed of lesss consequence to ascertain their quantity .
I did not , hovever , neglect this object of inquiry , but you wsill find that-some very curiox circumstances prevented me from attailling ite EXPERIMENrla I. Six pounds of dark grey melting cast iron rvere put into an eartherl retort ; a glass tube utas luted tv the neck , and its extremity was itumersed in water .
The retort was placed in a wilad furnace .
Before the retort and its contents could be supposed to be red hot , inflamlmable air came overv It burned svit ; h a deep blue flameg and was in no degree explosive .
It retldered lirne-wzatet turbid , azld was partly absorled .
When the retort had been heated about an hour and half , the air , hich was coming over pretty copiously , that iS9 at the rate of an OU1]Ct measure every three minutes , upon an average , suddenly ceased , and the apparatus , on examirlation , proved to be ; no longer air-tight .
Tlle retort ntas found to be cracked ; and the lumps of iron had rlone of therrl beela melted but they had been softened , and conglutinated together .
EXPERIMENT II .
Four oullces Troy of the same iron utere put into one of fMr .
WEDGWOOD &gt; S ear2cIlen tubes ) glazed and closed at one end.t Tilat t ?
nd of the tube was irzelosed in a barrelshaped crucible,0 the interstice filled with satld , and the crucibie reclilled soi dS to form a very small angle with-l the horizon : itz other respects the apparatus vtvras disposed as before .
011 the application of heat , air was again extricated , sooner vllan I shotllel llarre e2cpected , vf the same inexplczsive inflammable killed .
AbOtlt oneseventh of that which canze over first , and s^7hich traversed the water of the receiving vessels , was absorbed by milk of lime .
rrhe residue burned slowly , with a flame apparently not so deep as before the carbonic acid was separated .
Ixl this and the fortner experiment , the elastic fluids were most rapidly extricated on the first impression of a red or 0 white heat .
AfterwJards they callle over much more slourly ; during a considerable part of this experiment you might coun-t twelve slowly between e+7ery air-btlbble .
Whetl the utmost pon er of the furnace had been exerted Sor three hours , a phaerlomenon occurred wthicll produced some surprise in every person present ; and there were several who had been abulldantly accustomed both to chylmical and metallurgical operations .
A considerable absorption tooli place , and for ahout half an hour , it was necessary to blow air up the glass tube , toprevent the rater from rising into contact with the iron .
-It aftersvards appeared that the lead of the glazing was revived , :which sufficien-tly explains the absorption .
586 grains only of the iron had been completely fused .
The surface of tvo of the untnelted lurnps was curiously covered with numerous small blisters of metallic lead .
About seven hours after the fire was first kindledg it was discovered that the apparatus had iSailed .
I had examined the air that came over immediately before this acciclent , both hy means of lirrlewater , and milk of liLne , avvithout disc(3vering aIly vestige of carbonic air .
The iron weighed altogether three grairas more than at first .
But the adhering lead , and a quallti1 ; y of lead also svhicll Jas illcorporated with the iron , concealed a real , and probably a considerable , loss of weight .
The phnouletla it exhibited , WheIl pUt iIltO weak vitriolic acid , and the vitriolated lead which ssras formed , indicated the presence of this tnetal in all the superficial parts of the mass.\hen il ; had been kept some time in vinegar , it dissolved readily enollgh in vitriolic acid at first7 but the solution soon ceased , or became very slow .
EXPERIMENT III .
A coated flint glass retort was etnployed in this instancec The apparatus resisted a strotlg heat for two hotlrs ; and airs of the heavy inflamnzable kind catne constantly over .
E XPERI MEANT IX .
A coated retort of crown glass , containing six ounces Troy of the same iron , 7aS plaed OY1 a crucilble nearly -full of sand , and disposed as in the former esperilnetts .
I 1lCNv wished to measure the quantity of the alr and I theret8ore determined to receive it in merctlry .
It wtould have been 111 vain to attempt this in water , ola accoullt of the carlsonic acid air .
About tvelve ovclocli the retort was judged to be of a -dull red heat , and inflammable air came over .
The orifice of the transmitting glass tube vstas now covered to the depth of half an inch with mercury , when the discharge of air il ; lstantly ceased : the lute seetned entire .
Some of the rtlercury being remoured , so as toleave just enough to cover the mouth of the tube , irnmediately the air issued agai in bubbles , a proof that the apparatus was entire .
the mercury was poured into the trougll agaira , and in an instant there was a cessation of air .
The mouth of the tube being uncorrered and a lighted paper applied , a blue flarne appeared and cotatinlled to hurn , so great was the quantity of air discharged .
lUle orifice of the tube ras one-tenth of an inch in diameter .
We found that this constant flame could t)e prodtlced at any time during three hours and an laalf .
When water was substituted in place of zuercury ) air issued slowly , and as if with difficulty , under a presstlre of five inches .
When only half an inch was left over the mouth of the tube7 small 1 !
3ubbles ascended freely .
Durig a considerable time I counted four , slowly , between each of these bubbles .
I did not collect above three ounce measures of air , and this contaitled carbolaic acid .
It was past four onclock when the apparatus ceased to be air-tightn and the fire had been kept as strong as possible .
rflle iron was most completely fused .
There was a good deal of revived lead within the reCort ; there were also Inany glosbules ira the neck .
Probably some broken flirlt glass had been added to the usual tnaterials for crown glass ; I cannot otherwise account for the appearance of the lead here .
In the last experiment the lead of the flint glass had been revived .
EXPERIMENT V. Two ounces of the same iron , immediately upon being taken out of a retort , in which they had been kept , at a red heat , for about aa hour and a halE , anc } whickl &gt; tere therefore as free from s-vater as iron ca-l easily be procured , were put into ala earthetl tube , unglazed , and closed at one end .
This tube was clisposed as in experil:nent II , only the end of the glass tube was imnlersed in mercury , instead of water .
You will not be surprised to 11ear that air did not now come over so soon as in any former instance .
When the fire was raised to its full force , exactly the same amusing variety of appearances took place as in the last experiment .
Under the pressure of half an inch of mercury , ilOt a particle of air sZas discharged ; but the moment the pleasure +vas diminished to a small fraction of an inch , the bu-bbles succeeded each : other pretty quickly ; and so on repeatedly .
UpOll lowering the surface of the lnercury , and pouring some water upon it , I received more there two ounce measures of air , which , by the test of li-ne-w-ater , did seem to contain a vestige of carbonic acid , but it was too minute to be appreciated .
This experiment with the air vas made after a strong white heat had been kept up for t}ree hotlrs .
Soon aftersrards the bubbles ceased ; but +ve could not then , nor upon ) exami1ation of the apparatus when cold diSCOVer antr fai1Ure .
tithe ISre WaS St ; l1 kePt UP fOr three hOI1rS .
THe tUbe mUSt haVe been eXPOSed 0a STrOng B7hiTe heat seven hours in all .
Tlle iron had lost eleven grairas in weight .
( )thy about one half had been thoroughly fused .
The surface of two lumpss that had not been fused- , had the close texture , and silvery appearanceX of malleable iron .
The th-in edges yielded to the strolke of the harnmer , and a gentleman , perfectly collversant itl the nature of iron , agreed with me that it had all the characters of mallecable iron .
EXPERIMENT VI .
Thirty-one grains of -artificial plumbago , ill shining flakes from the iron works , were exposed in a small retort to a stroIzg -heat , for six hours , in the same pneumatic apparatus .
It was difEcult to separate , even by the help of the magnet , all the intimately missed particles of iron , and there were also a few particles of coak incorporated with the plumbago .
Air , of an explosive inflamlnable kind , was extricated , and rose freely through five in-ches of mercury .
We had not been sufficiently careful to let the lute fix before we commenced the experiment , and it soon failed .
Upon taking oWthe pressure of the mercury entirely , and repairing the lute as well as we could } we had every reason to believe that the air soon ceased .
The air received in the mercury contained oneeighth of carbonic acid The remainder exploded .
The plumbtago lost fbur Erair1S .
Mr. PELLETIER if I remember right , founal fthat native plutnbago , exposed to a fierce and long continued heaty lost to grains in 200 .
In the present experiment , its appearance was unaltered .
Probably the loss was outing to moisture imbibed by the particles of coak , and to a s1nall combustion by the air in the retort .
It will , I think , be admitted that these experiments abutzdantly confirrn the inferences I had formerly drawn from appearances by their nature less decisive .
The real extrication of air , varying in its nature at variotls periods of the process , seems to be placect beyond doubt .
The experiments in glazed and glass vesselsa were made with a view to exclude the possibi lity of the supposition OI the air entering through the pores X think that Dr. PRIESTLEYN if he should repeat these experiments.^ and find that they have been accurately made , will with his accustolned openness to conviction , abandon an opinion he has for some tirtle entertained ) and no longer consider water as essential to the constitution of elastic fluids .
Several observations might be made UpOll this point , and those which I have just noticed above ; but they 7ill readily occur to persons conversant in chemistry , and it is not the object of the publications of the Royal Society to teach the elements of science .
I shall , therefore , confille myself to the unexpected and anomalous appearalacesn and then attempt to draw a few useful *P uterences. .
I was stlrprised at tan-e extrication of inflaanm-able air in such low degrees of heat .
You have seen that cast irony highly charged wzith charcoal ( the pblogisto onustum of BERGMAN ) yields air at the tetuperature of melting lead .
For undoubtedly the blisters of lead , which lay upon the lrotl are to be considered as air bubbles caught in a solid film of lead .
Perhaps white cast iron would not yield air so readily ; possibly iron lzolds i-ts charcoal with nor forceas it contaills less * &lt; lnterlc W to malie SONle comparati &gt; re experim.etats UpOll the va rieties of cast iron , but so curious -an appearance as these blistez s will always be rather the bounty of acci(leint than the eiect of skill or labour .
-The obvicSus method to produce .
them +^rould be to cover the iron with lead .
All the unmelted luinps of irorl were porous , and operz ill their texture .
2 .
I anz at some loss how to explaill the occasional discharge and cessation of air , in one experiment in which a crowla glass retort uTas used and in another vlth an unglazed earthen tube .
There was no flaw in theluten nor in the vesselsn for it vas sdischarged for the space of several hours utader a small pres sure .
Either ) then , it utas forced througla the softened glass in the first and the dilated pores of the tabe ill the second case ; or it was absorbed by the substance of the vessels- ; or it was not extricated fkom the iron Of these suppositions t}e third seems to me the most probable .
It is not likely that an hole should be made through the melted glass urlder the pressutre of the half and closect under that of perhaps the eighth vf an inch ; or that pores in the tube shollld open and slout in con fornnityto such a variation of circumstances : and with re. .
guard to the tuben there can be no question as to absorptiono One principal difliculty as it appears to rnes in the matlufacture of iron is to get rid-of the charcoal.* sThe oxygene readily enough tInites with a small portion ; but the attraction of the iron on the one harld , and on the other , the little dis position of the charcoal to put orl the elastic form ill comparison with many other less fixeel substances together fortna very considerable obstacle to the change of charcoal info air ; and as I lnave already observed , the iron probably }olds the charcoal more strcngly as its qantity ( liminishes .
In this state of thing ; s a small adelitional impedilment will prevent the heat from throwing the charcoal into the state of air ; and some degree of pressure maIst be ade\#163 ; luate to this eSect * and why may not this points from svIliCEl as you recede on Qp posite sides , the attraction of the particles of charcoal for one another5 or for iron , either shall or shall not be overcome lDy heat , have been found in these ; experiments ?
The next COllSideration will both illustrate and confirtu the.se ideas. .
A chymist , those nOtiOllS Of iron are derived principally from books , and from the phxrzonerla which are ) presented by prencesses not having metallurgy for their immediate ob Ject , will be apt to consider sotme things related above as irconsistent : the violence of the heat , for instance , and the smallness of its eSects ; since even cast iron was not fused in all the experiments .
The fact iS7 WheIl cast iron exposes alarge surface , and heat is gradually applied ) it proves almo-st as infiasible as malleable iron : indeed , by the gradual action of heat it is converted , -^superIScially at least , into malleable iron , or approxirxlates towards it : and consldering oilly iron and charcoal , I believe , the fusibility of iron will be directly as theqaantity of ckfarcoal it contains .
Novx iI3l the experiments I have described ) pieces of one , two , and three drachms , and sometitnes less , were vlsed , forslarger could not be inserted into the rleck of the retor .
And , ill order to avoid this inconvellience in future I would recommetld cylirz(llers to be cast , of a diameter sait-ed to the mouths of the vessels .
This infusible coat would 1)e an impedimexat to the conversiotl of the parts below by pressing UpOll them ; the elastic fluids could not either traverse the solicl surface so freely as a liquid , azad perhaps , as I aln disposed to believe , they could not traverse it at all .
The malleable skin seerns close in its texture and the porosity of the rest might arise from the generation of just air enovgh to prodllce an i]teral expallsion .
Il1 the puddling operation , it is of the most lnaterial consequence to keep the mass in constant agit-ation .
Thus the parts are thoroughly blelldeds the attraction of cohesion is a good deal counteractetl , and there can be no pieces llide-boun(:l , if I may so explsess lmyself .
T1lis last , perhaps , is the greatest advantage derived frorll t1we labour of the workman .
4 .
I was asked by one of the most ingenious and proSoulad philosopllers of the present ag , e , why I had 1leglected the actiOll of the atmospheric air in the theory of the cotlversion of iron ?
It is simply because its action upon the metal seelms in practiG-e pernicious ; I consider its presence as an evil , though a necessary one , according to the present modes of workizg ; I was also ansious to try this opinion by the test of experinent ) and you see it has been fully confirmed .
In the last -experiment , part of the ircon wwas coluplet:ely collverted , aladin some otilers it seemed approaching fast tosvards nat^re , as the lnatlufacturers express it .
It is , ixldeed , very possible tc ) con-ceive a way in which air might be beileficial ; that is , if it -collld be applied $o as to burn the charcoal lmerely ; but at present , for one grain of charcoal which it converts into carfbolic acid air , it Gollverts many of iren into flnery cinder ; and as I have formerly shown , this is not the way in wthich iron is actually converted in the leverberatory , and probably laot in the finery , furnace .
t. It is impossible to ascertain the principles of any art- &gt; without immediately improving the practice , or capening a prospect of future improvement .
The preceding observations may serve to direct attempts to render the meLallurgy of iron less clifficult laborious , and expernsive .
For , z. if a quantity of oxygene , nearly sufficient to burn the charcoal , could be chyxnically combined with the cast iron , the operation Would constlme-less fuel , and would not require so long a time .
It may be worth while to consider if the ores of il on9 colltaitling snangalzese , owe any part of their value to this circumstance o. If it could be contrived to apply a sufficient heat tog large quantities of iron in close vessels , and at the same tinle , to agitate therl^l suicientlyX the loss in collversion nZould notS perhaps , exceed ten in arl hundred .
03- .
T}le inzportallt object of converting Brit ; ish iron illtOsteel , may possibly be attained byfollosving up reflections suggested by the foregoitig experinents .
When the oxygene has been separated in the form of carbotlic aci(l- , there will rernain the charcoal and irol , the constituent parts of steel .
Perhaps the materials , at a certain pe riod of the process , may be so rlea-rly approaching to steel as to be easily convertible .
The tnass will contaill also a quantity of sulphur , otl wlwich perhaps the difficulty of making good steel from our iron d-epends .
But this difficulty , I am persuaded , will IlOt be insuperable. .
It zuay be proper to add , that whenever attention was paid to it , the hepatic smell in the extricated air was perfectly disting ; uishable .
I hope youzwill also permit.lue to add , there ; whatever infbrtnation or advantbage may be derixTed from these facts and ObSeRVatiOnS , mUSt be in a greaT meaSUre aSCr ; bed tO the 1ibera1 CUriOS ; tie Of WILLIAM REYNOLDS , MThOSe en-erPriSing spirit and illventive genius have improved our machinery , enlar ; ed our nzanufactures , and changed the face of a large distt ict in his llative COUIlty .
I have the honour to be , &amp; cd ShiEnal , Shropshire , Bebrua3ry 14th , I792 .
THOMAS BEDDOESo P. S. The reSidUU1n Of 486 graiI1S OF CaSt irOn , the Same aS that USed in eXPeriment Ie Weighed 482 grainSN aFter beinS diSSO1\red in Weak VitriO1iC aCid , and heated tO a dU11 red heat ; the SaiMe qUanTitY Of ; rOn , after the eXPerimEnt , affOrded a reSidUUm Of 39 gra ; nS , and a 1itt1C mOre ; iX1 the reSidUUm left bY eqUa1 qUantitieS Of irOn , befOre a]]d after the eXPeriment ; n the Ung1aZed tUbe , there WaS a differenCe Of fiVe grainS ; the $01UtiOn Of the irOr1 that hAd been SUbmitted tO the eXPeriment Went On VerY S10WIY ; and WOUId not haVe been eSeCted by vitriolic atid in man ; y months .
In the latter case , I used sozle snuriatic acidX hich rquickly dissolved it : in the former , veak aqua regia was used for the solution of a very small part of the whole lump .
I suspected lead to have caused the slowness of the solution in the first case , but there carl be no such suspicion in the second .
TIle differelace between these resdaums tends to show that pluinbago was consumed by the heat : but they do not show the loss accurately ; for in the re siduum of the iron that had been fused itl the first experiment there was a small qUalatity of vitriolated lead ; and in the other there was besides lie plumbago a:small quarltity of thatfidifficultly soluble calx of iron , wheh the solutiorss of this tnetal deposit on long exposure to air .
The diSerence was greater therefore , than -it gppeared .
On the other hand* , the long action of the atid-ws might have consumed some plumbago .
There was little or no cals attractable by the magnet in the residuums of the fused irollo From the 48 grains of residuum:s I separated more than six grains by the-magnet .
tIDCtSCIt

