THE PREPARATION OF GUANID'ENE ETC. 1133 CXXVIL-The Preparation of Gua.nidine by the Interaction of Dicyanodiarnide and Ammonium Thioc yanat e. By EMIL ALPHONSE WERNER and JAMES BELL. BY heating a mixture of dicyaaodiamide and ammonium chloride a t 195O folr ten minutes Bamberger and Dieckmann (Ber. 1892, 25 545) obtained diguanidei C,H,N,. The yieild wa,s pooh and Ostrogovich (Bul. Soc. qtainfe Bucaregti 1910 19 641) using ammonium iodidei in plaae oif the ohloride under similar wnditiolns, obtained it much btter resulti. The reactioin has been represented thus: HN:c(NH,)*NH.CN + NR,*HX= HN:C(NH,)~NH*C(NH,):NH,HX. The dicyanodiamide complelx is supposed to remain intact whilst the! cyazmgen group unites with ammolnia. On this assumption the relaation is considelred to uphoild the cyanoguanidine formula for dicyanoldiamide propoeeld by Bambeirger (Ber.1883 16 1459). I n the statia oonditiotn dicyanoidiamide is stable up ta 2 0 5 O , when i b melb and is simultaneously depolymelrised t o cyanamide* and largely re~pollymelrised ta melamine (oompa(re Werner T. 1915, 107 715). Several experiments have sholwn that dicyanoldiamide is reladily depolymetrised to cyanamide at' comparatively low temperatures in the premnce of certain relagentis and it6 behavicrur in the presence * This fact is not referred to in the literature if a gram or two of dioyano-diamide i8 heated to the melting point in a porcelain crucible and quickly covered with a glass bell-jar standing over a layer of water the latter after a few momentswill be found to give quite a copious yellow precipitate of silver cytmamide on addition of ammonio-silver nitrate solution.u u 1134 WERNER AND BELL THE PREPARATTON OF CUANIDINE od ammonium thiolcyanah folr example( furnishels a case in point whiah is olf pradiaad value. P'urei guanidine t,hiocyanato has been easily prepared in amord-mm with the squatlion C2H4N4 + 2NH4*SCN = ZCH,N,*HSCN . . . (1) Whilstl the main change is equa81 to 90 pelr c,entl. od tlhei theoretical, it. is aocrompa,nied by a seloonda,ry rea,cltion whereby thi~amme~line is formed in small quant'ity. This does not. interfe're with the sucaelss of t,he prelpa8ra.tioln as its sepa,ra,tio(n from t,he chief proiduct is a simple rnaf'br. Pure diayanodiamide wa8s useld in preliminary sxpelrimelntis tIo detefnninei the b.elstj conditioas but' from an eloonoimia point of view the commercia,l ma,terial may be .conveniently employeld.The1 sample use'd in the following prepasat,ion contained dicyano-dia,mide = 95*5,* oa,lcium casb(o1nat.e = 3.25 mellamine = 1.25 per cent. An intima.b mixture of 43.5 grams oQ dicyanoidiamide (=42 grams pure) aad 76 grams olf dry ammonium t,hiocya,nafei in coarse polwder wa,s hela8iad in a tall na'rroIw beakelr past,ly immelrseld in a glymlroll-bath. A 10,ose cIa,rdb,oa,rd oovelr ca$rrie8d a the:rmolmelter, which served at the same1 tlimel as a stlrres. At' about' 80° tlhel mixture began to1 mellt,; the i;ernpelra.tare wa.3 gradually raJsed to1 120° about selventy minutm being required for t,he purpose and this tempe8ra,t.ure was maintaineld for t-hree and a,-ha,lf hours.As the resct8im prolceode,d the product which in the earlielr s t a p was an almorjt clela,r liquid graduadly betmms very visoous. It was notw t'reahd witholut previous coo,ling with a,bout 250 C.C. o,f wates aad a,llowed t,ol digest unt'il cold. After filtrat'ion and wa.shing, 7.4 grams of amolrFhosus relsidue welrei separa<teld (residue A ) . The filtlra,te was colnclentlrated a8 far a,s possible by eva,polra8t8ion a.t loo", aad the ciystladlinel mass which formed on cooling weigheld wheln dry 110 grams (Folund SCN = 47.76 ; guanidine thiocyanat'e relquiresl 8CN=49.15 per oent.). The product wa.s redissolved in aboat 200 0.0. of warm wafer and 1-02 grams of amorphous solid (residue B ) welrel separateld by filtra.tion.The filt8ra4tej wa,s e:va,porateld t'ol a' syrupy oolnsistency a t 10Qo and a,fter " seelding " thei cold solut.io,n with a minute cryst,al of guanidinel thiocyanat'e itt quiokly set to a. crystlallinel mass of thO pure1 salt (Found SCN=48.96 pe.r cent,.). The yield of pure A single recrystallisstion of commercial crude dicyenodiamide is The material generally sufficient to obtain a product of this degree of purity. should be free from calcium hydroxide which is occasionally present BY THE XNTERACTION OF DICYANODIAMIDE ETC. 1135 guanidinel tihiocyanatel was 107.1 grams equal to 90.8 per cent. of the theoretical. The oharacteristlic picrate was prepared a.s a means of idelntifi-oation. Whilst residue A colntained the impurities originally present in the1 dioyanodiaanide 5.4 grams of pure thioammeline, C,TT,N,S weire separated from it1 by trelatmetnt with sodium hydr-oxide sollutioln and precipitation after filtratioa by carbon dioxide.It was identified by its properties as deswibeld by Elason (1. pr. Chem. 1886 [ii] 33 290). The picrate melted ak 215O and as it separated from solution in peculiar crystalline formations which, under the) microscotpel resembled a highly divided palmatifid lelaf, this colmpound may be used for the idelntification of the substance. Since thioammeline results from the interaction olf dicyanoi diamide and thiocyanic acid. according t o the equation i t is not! possible to supprms it's formattion a8s dissolciation of ammolniurn tlhiocya,nalte to a smasll extsnt ca.nnolt.b~e avoided under the conditions' of the1 elxperiment. Residue1 B was pure thioammellinel geaerabed as above from small quaa t,it,im of dioy anod iamidel a,nd amin olni um thiocy an a,te which ha.d escaped change. This warns proveld by eivapora.ting to dryness a t looo al so3utlion od 2 grams of dicyaaoldimide and 3-6 grams of almmonium tlhiocyanate in 50 C.O. of water wheln 0.29 gram of thio-amemline was obkined. For this relasoa e.vapora#tion of the solu-tion t,o dryness in the first inst,a.noe wa,s found t40 be1 the most. efficient melt4hod for pmpasing a pure prolduct and the1 a,ddition of ab,out a gram of ammo,nium thiocyana,te t'ol t'hs fusion when the1 remtiotn haa been in progretss €or t'wol hours is a8dvisa8ble. When a.n 0xw1ss of the salt was used at' the outlset of the experiment the yielld of thioammeline was increased.Shoald the cryst,als of guanidine thio,cya,na.ts sho'w opalescence after arystdlisatlon from and while strill in hhe mo,ther liquor a, second sotlutim and filtra,tion is necessary t'o remove the la.& traces of t,hioammelinel. On a.ccoIunt. olf the grelat solubility of guanidins thiooyanate in water its complete separatlion in crysta,lline form is tedious; it may be remvered as carbonate by mixing the visoous mo,theIr liquor with ab,out four volumes of alcohol and afte'r the a,ddit,ion of the requisite amount of potassium hydrox.ide t.he guanidine is precipit'ate.d by a current od casbon dioxide,. Apa,rt from itls pract'ioal va.luel the re,a80tion is of t.heore,t,ical interest since i t is obvious that delpo,lymerisat,ion of dicyanodiamide u u* 1136 THE PREPARATION OF OUANIDINE ETC.must be the first! pha\se of the ohange in order ta yield guanidine in such quantity as relquireid by elquation (l) whilst the foirma,t,ioln olf t,hiolammeline the1 mastlitutioln off which has been cle8arly dmoln-stsateld by Klasoln (Zoic. cit.) acoolrding tho equahioln (2) woald be difficult t o elxplain on t'hhe basis of the '' clyanoguanidina " structu rc of diuyamdiamide. Considering t~ha.t Bamberger and Dieakmann (Zoc. cit.) obt'ained digumide by heIat,ing guanidine hydrochlolride a.t 1 8 5 O thelre can be no doubtt tlhat the lattler was first formed in their prepa,rtrt.ioln of diguanide from dicyanoldiamide. This has been verified from qualit,amtivel egxperiment~ which h.owever welre olf no1 praoticd interestl; on the1 other ha,nd very encouraging rwults have been abtaine'd in the prepasation of methyl- aad ethyl-guanidine an a,ccolunt od whioh i5 reiselrveid far a future communica4tion.Sol far a5 the a.uthors a<re a,wa,re nolne olf the methods hitherto demribed for the1 preparation of guanidine can colmpare with tyhO process now sat forkh either in mspe& od cost, simplioitly or yield oif pure produot,. I n this colnnexioln it may be useful t'o relcoiUllt the autholrs' experience of Ulpiani's method (1909 D.R.-P. 209431) foir the prelparatioln s f guanidine nitrate by the acljioln of amqua regia at 60-65O on dicyanodiamida. A quarntit8ative yield is daimed. Since the change (indireot oaxidation) is accompanied by the e;vollut.ion of muah ca,rbon dioxide and nitmgela from dicyanodia-midine simult,aneolusly produced the maximum yield of guanidins is represented by the ratio C,H,N -+ CR,N3,HN03; from three casefully conducted experiments the beat yield was 26 grams of guanidine nitrah from 42 grams olf pure dicryanoc diamide equal ta 42-7 per mint.of thO theoreitiaad calculated on the folregoing ra,t,it.ia. It is evide,nt from this result t,hat not less than two1 mollelmla od diayanodiaznide asre invoilveld in t.he production of olnne of gua.nidine bsy t,his method. The assa,y of dioya'noldiaxnide was colnveniently cronduoted by hydrolysis aa follows 2 grams of &he powdereld sample were mixed with 25 O.C. of water N-hydrochloric a,aid was addeld until the alkaline impurities ware neukalised using melthyl-orange as indica,tor after whioh the pro,duat was heated for fort.y-five minutes a,t looo with 25 O.C. of N-add. The residual acidity was deter-mined by titaa,tion wit,h N-sodium hydroxide. Earoh 0.0. of N-acid neutradised is equal to 0.084 gram of diayanodiamide. UNCVERSITY CHEMICAL LABORATORY, TRMlm COLLEUB, DUBLIN. [Received A q w t 24th 1920.