I THE POSITIVE CHARACTER-, o f the
HALOGENS JACOB KLEINBERG University of Illinois College of Pharmacy, Chicago, Illinois
T
HE student in general chemistry is taught that the halogens are typical nonmetals and that their abiiity to form inorganic compounds depends upon the ease with which they pick up an electron to complete their outermost orbits. The electronegativity of these elements is continually stressed, and i t is pointed out that the chlorine atom, having the greatest affinity for electrons, can displace bromine and iodine from solutions of their salts. Insufficient emphasis is placed on the fact that this is true only when the halogens are acting as negative elements. Little or no mention is made of the fact that in the oxyhalogen compounds, the reverse is true--namely, that iodine will replace chlorine from potassium chlorate or bromine from potassium bromate. The discovery in recent years that iodine and bromine may form coordination compounds with organic bases, such as pyridine, in which they act as positive univalent ions, will undoubtedly initiate new investigations. Therefore, a discussion of some of the aspects of the positive character of the halogens is in order. COORDINATION
The following were prepared as definite crystalline compounds: 0
/I
I(PY)s.NOI,I(fiy)NOa, I(Py)*.C10a. I @ Y ) ~ - C - C H I ,
COMPOUNDS O F POSITIVE UNIVALENT IODINE AND BROMINE
The suggestion of Cofman (I) that hypoiodous acid may have the properties of a very labile base, IOH, is brilliantly substantiated by the investigations of Carlsohn and his students (2, 3, 4). Reasoning that IOH might be stabilized by coordination with pyridine (py), these investigators succeeded in preparing a series of salts of the hypothetical bases I(py)OH* and I(py),OH. From their behavior, there is no doubt that the iodine exists in a positive univalent state in these compounds. The salts are generally prepared by treating the silver salt or the mercurous salt of the necessary acid with the calculated amount of iodine and a slight excess of pyridine in some nonaqueous solvent such as chloroform. Reaction occurs immediately according to the equations
0
+
2 1 ( p y ) - - O - - ~ ~ 2 HgI3.2 fiy * The I .py+ ion has been postulated to account in part for the conductivity of iodine in pyridine [Audrieth and Birr, J. Am. Chem. Soc., 55,668 (1933)l.
The chemical reactions of these salts show that the iodine is positive and univalent. When dissolved in sodium hydroxide containing potassium iodide and then acidified, they all liberate free iodine. I+
+ I-
-Iz
This reaction is utilized for the determination of iodine in the compounds. With potassium chloride and potassium bromide a double decomposition takes place with the formation of ICI.py and IBr.py, respectively. The iodi-pyridine salts hydrolyze slowly in water in accordance with the following equations:
-+
I(py)t.NOs
5 I(py)OH
+ +
HOH + I(fiy)OH fiyH.NOr 2 Iz pyHIOn 2 HOH 4 py
+
+
Iodi-pyridine salts react immediately with an acidic phenol solution to form iodophenol. Cofman ( I ) has presented evidence to indicate that iodination of phenol is accomplished only by positive univalent iodine and not by the iodate ion, free iodine, nor by the iodide ion. By treatment with sodium hydroxide solution the bases I(PY)~.OHand Ipy.OH are liberated. These immediately go over to their respective anhydrides,
ing to the equation [I?] Pr [I(py)z] X and thus should become conductors. The results of the conductivity experiments are given in Tables 1 and 2. From Table 1 i t is apparent that although Ipy.NOs has a small equivalent conductance in acetone in comthe anhydride of IOH, can be stabilized by coordina- parison with I ( ~ Y ) ~ Ni O t still ~ , conducts the current aption with pyridine. preciably. Carlsohn makes no explanation for this On the basis of the decomposition potentials of ICl, phenomenon. It is possible that acetone enters the coIBr, and IC13, and electrode potential measurements of ordination sphere to some extent and fonns a coordinate the iodine electrode against solutions of the above covalent bond with the iodine by means of a free pair of substances, Finkelstein (5) has demonstrated that the electrons on the carbonyl oxygen. This would drive I+ ion should be placed with the noble metals in the nitrate ion out of the coordination sphere and thus make electromotive series. This was substantiated in a quali- the compound a conductor. The data in Table 2 lend tative manner by Carlsohn (2). A chloroform solution credence to such a mechanism. Monopyridine iodine of dipyridine iodine (I) nitrate was treated with metal- (I) nitrate conducts the current in methanol equally well lic zinc, magnesium, iron, copper, silver, mercqy, gold, whether or not pyridine is present, whereas the conducand platinum. All replaced the iodine and were imme- tivity of the benzoate is increased tremendously by the diately dissolved. However, it is impossible to say that addition of pyridine. It should be borne in mind that iodine is more noble than silver since dipyridine io- the nitrate ion is an extremely electronegative unit and dine (I) nitrate was prepared by the replacement of it would therefore exhibit a lesser tendency to form a silver with iodine. Evidently in the case of silver the coordination bond than the benzoate ion. In the case reaction can go in both directions. In each case the of the nitrate, methanol may well coordinate to the posiinsoluble silver iodide formed removes the replaced ele- tive iodine through a free pair of electrons on the oxyment from the reaction, and perhaps this accounts for gen. On the whole the conductivity measurements the apparent reversibility. leave considerable doubt concerning the coordination Electrolysis of dipyridine iodine (I) nitrate in water- number of positive univalent iodine. pyridine and methanol solutions yields iodine a t the Salts in which methyl-substituted pyridines have cathode, while the solution around the anode remains been coordinated to positive univalent iodine have also clear. However, it is entirely possible that the iodine been prepared by Carlsohn (3). These are [I@-picois deposited a t the cathode because of a secondary re- line)z]N03, [1(2,6-1utidine)2]NOa, I(2,4-lutidine)zN03, duction. To rule out this possibility, dipyridine iodine and [1(2,4,6-c0llidine)3~]NO~. (I) nitrate was electrolyzed in anhydrous chloroform, The corresponding compounds containing positive and here again iodine was precipitated a t the cathode. bromine, [Br(py),]N03 and [Br(py)z]C1O&, have also This proves beyond doubt that the iodine is the positive been prepared by Carlsohn (4) by methods similar to constituent of the salt. those used for the preparation of the iodine salts; the Assuming a coordination number of two for positive former was first prepared by Uschakov and Tchistov univalent iodine, Carlsohn postulated the following (6). The method of analysis of these compounds demtypes of structure for the salts: [ I ( ~ Y ) ~ ]and X [I?] onstrates that the bromine is positive. A sample of a (X = acid radical). An attempt was made to verify bromi-pyridine salt is treated with potassium iodide disthese structures by conductivity measurements. Salts solved in sodium hydroxide, the solution acidified with of the type [ 1 ( p ~ ) ~ ] X should be good conductors, sulfuric acid, and the liberated iodine is titrated with whereas those of type [I?] should not conduct the ~ l e c - sodium thiosulfate. One atom of bromine liberates two tric current. Moreover, salts of the latter type should of iodine. revert to the form [ 1 ( p ~ ) ~ in ] Xpyridine solution accord(Br+ + 21Br- + I3
which may be isolated.
This indicates that
-
THE ELECTROLYSIS O F THE CYANOGEN HALIDES
TABLE 2
BamvaLelrr Comnmmc~.OF Ioor-P-ms
-
Dilulion V
100
Sd1
Surs IN MslaaaoL br 20'C. Ohm-1 cm.*
IPY.NOI Ipy .NOx about two mols of pgridine IIY.OOC.QHB Ipy.00C.CaHs about two mols of
+
pytidine
+
55.3 55.3
2 15.2
There is excellent evidence that the iodine in such compounds as iodine monochloride and monobromide is positive. In each case the iodine is in combination with a much more electronegative unit. Orton and Blackmann (7) found that both of these compounds on hydrolysis give hypoiodous acid, which undergoes a self oxidation-reduction to iodic acid and iodine. In the cyanogen halides-chloro-, bromo-, and iodocyanogen-the halogens are in combination with a halogen-like group. Nef (8) found that bromo- and chlorocyanogen yielded potassium bromide and chloride, respectively, when boiled &th potassium hydroxide.
whereas iodocyanogen formed potassium iodate and iodide. From this one might infer that the bromine and chlorine were negative in these compounds while the iodine was positive. It has also been shown (9) from the action of hydrohalic acids on iodocyanogen that the iodine has a positive character. The reactions cited above may be summarized in a series of equations.
to be the normal iodate of trivalent positive iodine, I(IO&. The same compound has also been prepared by the treatment of iodic acid with phosphoric acid (12). The basic iodate, 10(103),is produced by the action of hot, concentrated sulfuric acid on iodic acid, followed by treatment of the product formed with water (11). The oxidation of iodine by perchloric acid or by ozone and perchloric acid gives rise to the formation of &EN + 2 KOH KOCN + KBr + H.Ot (1) a perchlorate, the composition of which may be represented by the formula I(C104)3,2Hs0(13). The normal sulfate, I2(SO& is formed when the basic sulfate is +treated with sulfur tr~oxidea t 100-120°C. ICN + 2 KOH KCN + KO1 + HzO (3) Fouque (14) first discovered that fuming nitric acid 3 KO1 KIOa + 2 KI will oxidize iodine in the presence of acetic anhydride to +ICN HCl HCL ++lCl the normal triacetate, I(OOCCH3)s. Fichter and Stern (4) (13) utillzed this reaction to prepare a whole series of Clark and Streight (10) studied the electrolysis of the compounds of trivalent iodine. The results are sumcyanogen halides in various solvents. From Nef's work marized in Table 4. on the hydrolysis of the cyanogen halides one might have expected that the iodine in iodocyanogen would migrate to the cathode in all solvents. Such, however, is not the case. As is seen from Table 3, only in pyriCampound Method of prcporotion IPO 1, + HIPO~+ (CH;CO).O +eoncentrated HNOI dine does the iodine react electronegatively.
+
--
-
TABLE 3
Solulc BrCN
d~poril dcPori1 Cyanide Bromide Bromide Cyanide Bromide Cyanide Bromiae cyanide Noneonductor Tarry precipitate ~ r ~ m i d ~ Cyanwen gas Iodide Cyanide sodide cyanide Iodide Cyanide Iodide Cyanide
Solvent
Acetonifrile Nitrobenzene Methanol E ~ ~ Y glycol I C ~ ~ Puran Pyridioe ~ni~i,.~
ICN
Benzene ~cet~nitrile Nitrobenzene Methanol Ethanol ~ i t ~ ~ Pyridine Aniline
.
++CChCOOH + (CHICO),O + fuming HNOI CHChCOOH + (CHS0)sO + fuming
I(CO0CCla)r I(COOCHCI&
13 It
I(S0zCHd.
I(C0OCHs)z CHJSOIH L CC1,COOH 0 8
+
I(COCCla)r.I(IOa)r
+
+
These compounds are sensitive to moisture and quite unstable to heat. They hydrolyze readily in accordance with the following equation: 5 IPO. 9 HnO-11 + 3 HIOJ + 5 HsPOI
+
.
The fact that the iodine in the compounds described above exists in a positive trivalent state has been strikIodide . ingly demonstrated by Fichter and Stern (13). When a cy anopen gas and paracyanogen saturated solution of iodine triacetate in acetic anhyQuinoline Iodide ... dride is electrolyzed, iodine is deposited on the cathPyrrol Reacted chemically ode in direct concordance with Faraday's Laws. Positive trivalent iodine may also exist as part of the .Cyanogen iodide seems to behave as an electromer, that is, it may ionize reversibly as follows: anion in polyhalide complexes. The acid, HICla4Hz0, is obtained when iodine, suspended in concentrated hy.+ I- . C ? .-~- N drochloric acid. is treated with chlorine 115). The - - % =salts, KIC14, ~ 6 1 CsIC14, ~ 1 and ~ M~{ICI&~H;O may Clark and Streight found that there is no direct come- also be It has been shown (16, 17, 18) that lation between the concentration of the two electromers the IC14- ion is one of the more stable of the polyhalide present in solution and the dielectric constant of the anions, and this stability is attributed to the fact that solvent. However, it is apparent here, as in other ex- iodine (a large electropositive ion) is by amples involving positive halogen, that the solvent atoms (small electronegative atoms). d. a y s an im~ortantand as vet undetermined role. ~
~
Iodide t~ ohd i d~e n Cyanide Iodide
Cyanide
~
cyanide
-
IODmE There is considerable evidence that iodine may also exist as a trivalent cation. Partington and Bahl investigated the action of oxidizing agents on elementary iodine and confirmed the observation that ozonized oxygeu will convert iodine to a compound having the formula 1 4 0 9 (11). These workers considered this substance t The charges shown indicate polarity. POSITIVE TRNALENT
LITERATURE CITED
(1) COPMAN, V:, I. Chem. Soc., 115, 1040 (1919). (2) C'LRLSOHN. H., "Uber Eine Neue Klasse Von Verbindungen des Positiv Einwertigen Iods," Verlag Hirzel, Leipzig,. 1932. (3) CARL~OH*, H., Anpm. Chern., 46, 747 (1933):
$1 gg,",",",",",.IE.i.,,,i; 3$tg&:f)iz4, I., 0. TCHISTOV, (6)
USCHAKOV,
(1935).
M.
(7) ORTON,K. J., 870 (1900).
AND
AND
W.
285 (1926), Ber., 68, 824
W. L. BLACKMANN, I. Chem. Soc., 77,
(8) NEF, J. U., Ann., 287, 265 (1895). (9) CHATUWAY, F. D., AND J. M. WADMORE, J. chem. Sot., 81, 191 (1902). (10) CLARK,R. H., AND H. R. L. STREIGHT, Trans. ROY.Sac. Cam& (31, 22,323 (1928). (11) PARTINGTON. J. R., AND R. K. BARL,J. Chem. Soc., 1258 (1935). (12) FIGHTER, F., AND H. KAPPELER, Z. anorg. Chcm., 91, 134 (1915). .
(13) FICHTER,F., AND S. STERN,Heb. C k h . Ach, 1 1 , 1256 (1928). (14) F o u ~ u s G., , Chem. Zlg., 38, 680 (1914). (15) EMELBUS, H. S., AND J. S. ANDERSON. "Modern Aspects of Inorganic Chemistry," D. Van Nostrand Company. Ine.. Piew York, 1939, p. 331. H. W., AND D. R. DUNCAN,J. C k m . Soc., 2243 (16) CREMER, (1936). (17) CREMER, H. W.! AND D. R. DUNCAN, aid.,181 (1933). (18) READE, T. H., %bid., 2528 (1926).
