June 20.

1964 TRIETHYLAMINE-IODINE COMPLEX 2347

show that the porphine skeleton of each molecule is dis-

torted because of close contact with phenyl groups of
neighboring molecules. These phenyl groups are di-
rectly above and below the porphine skeleton, as shown
in Fig. 3 .
Z n T P P , ( H 2 0 ) 2 on
q the other hand, cannot pack as
closely as C u T P P because of the oxygen ligands, which
are 2.33 A, above and below the metal atom. The
greater amount of void volume in the zinc porphyrin
unit cell is illustrated by Fig. 3 and 4 which compare
the Z n T P P . ( H 2 0 ) 2and C u T P P crystal packing in the
x- and y-directions. In the z-direction, the shortest
Zn-Zn distance is 4.86 A,,
whereas the shortest
Cu-Cu distance is 3.76 A. Since the Z n T P P . ( H 2 0 ) 2
molecules are farther apart, interference from phenyl
groups is lessened and the porphyrin skeleton can re-
main planar. F e T P P O H ' H 2 0 ,with the oxygen ligands
above and below the ring, packs in about the same way
as the Z n T P P . ( H 2 0 ) *and is thus almost planar.
T h e packing and molecular interactions of nickel
etioporphyrin T4 arc illustrated by the 001 projection
of isomorphous nickel etioporphyrin I1 given by Crute.'
There are no oxygen ligands above and below the ring, I
and close contact between neighboring molecules is e
possible. The close packing occurs with a slight loss of Fig. 4 -Projection of l l / z unit cells of Z n T P P . ( H 2 0 h on 001
planarity. T h e ethyl and methyl groups, which are The scale is the same for both the C u T P P and ZnTPP.(H?O)?
projections
attached to atoms 2 and 3 of the pyrrole ring, are bent
out of the plane of the ring. The interactions of these is quite flexible, since crystal packing forces can bend it
groups with neighboring molecules contribute to bend- easily. The shape of a particular porphyrin in a crys-
ing of the porphine nucleus. T h e interaction of over- tal appears to depend on the substituents and the type
lapping pyrrole rings from neighboring molecules may of packing. Porphine, since i t has no substituents,
also contribute to the nonplanarity. would be expected to be nearly planar. The X-ray
T h e evident conclusion is that the porphine skeleton analysis of this compound is in progress.

[CONTRIBUTION
FROM THE DEPARTMENT
OF CHEMISTRY, STATE
THEPENSSYLVANIA USIVERSITY, PARK,PEXNSYLVASIA]
UNIVERSITY

Molecular Addition Compounds of Amines and Iodine. Evidence for the Existence of a
2 : 1 Triethylamine-Iodine Complex
AND D. &I. HART
BY C. D. SCHMULBACH
RECEIVEDSOVEMBER
4, 1963

Calorimetric studies of the triethylamine-iodine system in n-heptane revealed t h a t the measured heat of
reaction of the amine and iodine t o form a 1 : 1 complex was strongly dependent upon the amine concentration
This result is explained in terms of the existence of a second iodine(1j complex, [ ( E t a S ) 2 1 +[]I - ] , which is in
equilibrium with the 1 : 1 complex. T h e molar heats of reaction t h a t accompany the formation o f 1: 1 com-
plexes were measured for seven amine-iodine systems in n-heptane by the method of thermometric titration
and found to cornpare favorably with those values obtained by spectroscopic techniques. The sensitivity of
iodine( I ) compounds t o hydrolysis and disproportionation is discussed.

Introduction Additional information germane to this study has been

Numerous studies have been made of the inter- obtained from conductance studies and spectral i n -
action of the halogens and interhalogens with nitrogen- vestigations. T h e increase in conductance of pyri-
containing bases.' The large heats of reaction that dine with the addition of i ~ d i n e ,and
~ . ~ of acetonitrile
accompany the formation of 1 : 1 complexes of alkyl- with the addition of iodine or iodine moriochloride,*,g
amines and and pyridine and iodine5 indicate leaves no doubt that ionic species are present. The
that interaction is appreciable. The trialkylamine- nature of the ionic species as yet has not been estab-
iodine c ~ m p l e x e sare
~ ~ ~formed with the largest en- lished with certainty. Spectral evidence indicates
thalpy loss (-12 kcal.) of all the iodine complexes for that in solutions of iodine or i,odine monochloride which
which thermodynamic constants have been measured. contain a Lewis base, there exists one or more type of
iodine(1) cation. Popov and co-workersl"I have made
( 1 ) See 1,. J. Andrews a n d R M . Keefer, Ailuan. I I ~ O Y EC ,h e m . Rndiochem.,
3 , 01 (l<I61), (6) L F Audrieth a n d E J Birr, ibid , 5 5 , 668 fl9:3:3)
1 2 ) H . T a u h o m u r a . J . A i n . C h e m . Soc., 82, 10 (1960). ( 7 j G K o r t u m and H. U'ilski, Z physik C h o n fI.eipzig). 202, :35 , I<ITi:ir
, : < ) H Y a d a , J T a n a k a . a n d S . N a g a k u r a , B i d . Chem. SOL. J a p a n , 33, (8) A I P o p o v a n d 1%'. A. Deskin, J A m Chrm. .SOL 8 0 . 2976 ilOT,8)
l(i(i0 (15160) (9) A. I. Popov a n d S E. Skelly, t h i d , 7 7 , X722 (14553
( 4 ) S S a g a k u r a . J A m Chem Soc , 80, 520 (1958). (10) A I P o p o v a n d R 1'.P R a u m , thiii., 79, 570 f1957)
(,-I) C . lipid a n d K . S. Mulliken, t b t d . , 7 6 , 3860 (1954). (11) A . I P o p o v a n d K €3 R y g g ibid.. 79, 4022 ( 1 H 5 i )
2345 C. D. SCHMULBACH
A N D D. A I . HART
& A
VOLUME OF TRIETHYLAMINE TITRANT
Fig. 1.-Tj.pical titration curves for triethylamine-iodine
sl-stem samples I and V of Table 11: h, start of recorder chart
drive; B, start of titration; C, end point of titration; C-D,
portion of curve corresponding t o excess titrant; A T , measured
temperature change.
spectroscopic studies in which the results have been
interpreted as evidence for the existence of a positive
iodine species coordinated to two pyridine or pyridine-
like molecules, or to a single dipyridine molecule.
The fact that iodine(1) can have a coordination num-
ber of two is confirmed by the existence of such com-
pounds as [Py21]C10~,'2 [PyaI]NOa,'* [ P y J I F , l 3[ P y J l -
SbFs,I3 [Py,I]SOaF,13and [(y-picoline)zI]I.14 The
last cornpound is the only 2 : 1 amine-iodine com-
plex that has been isolated. X-Ray radial distribu-
tion studies showed t h a t the iodine-iodine bond is
broken in this complex. Other workers6,15,16have
interpreted their spectroscopic data as evidence for a
solvated iodine( I ) species in which iodine is coordinated
to a singie pyridine molecule. Isolation of salts of the
type [PyIIX. [a-picoline.I]X, and [@-picoline.I]X
lends support to such a Hassel has questioned
the conclusion, based on spectral studies, that 1 : 1
complexes of amines and iodine contain iodine( I)
cations.Ig X-Ray analysis of the 1:1 complexes of
trimethylamine and iodine or iodine monochloride,
( 1 2 ) H. C a r l m h n , "Uber eine neue Klasse von Verbindungen d e s positive
Einvertigen Jods," Verlog von S Hirzel, L e i p i i g , 1932.
( U )H. Schmidt a n d H. h l e i n e r t , .4?2gew. C h e m , 71, 126 (1959).
( I t ) D I,. Glusker a n d A. Miller. J . Chrm. P h y s . , 4 6 , 331 (1957).
( 1 5 ) R A . Zingaro, C. A . V a n d e r W e r f , a n d J Kleinberg, J . A m Chem.
Soc , 73, SR (1051).
f l 6 ) J . Kleinberg, E Colton, J. SattiLahn, a n d C. A . V a n d e r W e r f , ibid.,
7 b , 4 4 2 11993).
( 1 7 ) R A . Zingaro, J . E . Goodrich, J , Kleinberg, a n d C. A . VanderU'erf,
i b i d . , 71, 5 7 5 (1949).
118) 0. Hassei, Mol. P h y s . , 1, 241 (1958).
Vol. SCi
and of pyridine and iodine monochloride, show the
N-I-X bonds are linear. The I-X bond is elongated
and the S - I bond is of such length as t(J suggest the
bond is essentially covalent. X-Kay analysis of the t : 1
pyridirie-iodine complex gave similar results. 1 ne-
spite these convincing arguments lor the nonexistence
of iodine(1) cations in solid 1 : 1 complexes. the effect
of solvation and dielectric of the media cannot be dis-
counted. and there remains the possibility that a
[Ani. I ] cation exists. albeit in low concentration.
+
Higher orders of coordination are possible but have
been given no experimental \-erification. Spectro-
scopic studies have provided evidence for I.c- i n solu-
tions of iodine dissolved in pyridine5,'6 or iodine in
quinoline.'6 The anion ICI2- has been observed spec-
troscopically in solutions of iodine monochlorides-io,I!)
and nitrogen bases; IBr2- was observed in iodine mono-
bromide solutions.'" T h e presence of iodide ions in the
iodine solutions or of chloride ions i n the iodine mono-
chloride solutions has been assumed by sexreral invcsti-
gators. Such an assumption is reasonable although the
concentration of these species is undoubtedly very
small.
Calorimetric studies carried out in this laboratory
revealed that the measured molar heats of reaction
for a number of alkylamine-iodine systems were
strongly dependent upon the concentration of the
amine. This information suggested t h a t more than a
single complex species was present in the solutions of
tertiary amines and iodine dissolved in "inert" sol-
vents and prompted us to carry out a more detailed
study of the concentration dependence of molar heats
of reaction.
Experimental
Reagents .-Baker and damson resublimed iodine was ground
with 25% by weight of potassium iodidc and lO<'i,by weight of
calcium oxide. The mixture was sublimed; tlie iodiiie was col-
lected and stored in a desiccator over phosphorus pentoxide.
Reagent grade carbon tetrachloride \vas shaken for s
hours with a concentrated solution of potassium Iiydroyitie.
separated, and washed with distilled water. This proces5 \ w s
repeated twice more. The reagent was then shaken with con-
centrated sulfuric acid, separated, and washed Ivith tlistilletl
water. After the reagent liacl been washed free of tlic acid, it
was dried by shaking over anhydrous U.S.1'. grade c:tlciurn
chloride. The dried liquid was fractionally distilled froni ;r frcsli
quantity of calcium chloride arid the constant-boiling ceirtcr
fraction retained.
Reagent grade n-heptane was shaken fur intervals of t i hr. with
concentrated sulfuric acid and the acid separated. The procediire
was repeated until the acid no longer became discolored. 'l'lic n-
heptane was then separated, washed with distilled water. a n d
dried overnight over anhydrous calciurn chloride. The tlried
solvent was fractionally distilled from fresh an11ydri)us c:rliiuin
chloride and the constant-boiling center fraction retained.
Comparison of the ultraviolet absorption spectra of the purified
solvents with Spectrograde reagents was the criterion for purity.
Technical grade pyridine was refluxed for several hours over
potassium hydroxide pellets and then distilled from fresh potas-
sium hydroxide. The amines were Eastman Kodak White Label.
These amines were purified by fractional distillation. The con-
stant-boiling middle fraction was retained. L-apor phase chro-
matography indicated the amines were pure.
Apparatus for Calorimetric Studies.-The apparatus \vas cs-
sentially that developed by Jordan and co-workers.20,21For wcrk
in nonaqueous solvents with small heat capacities it was found
necessary to introduce a simple, but importcrnt, modification to
the apparatus. It is essential in this procedure that titrant a n d
titrate be near tlirrrnal equilibrium prior to titration. =\I1 at-
____~_
(19) R. E B u c k l e s a n d J I: Slills, J A m Cheipi .SO< , 76, 4845 ! l Y X
(20) G . J. E w i n g , P h I ) . Thesis, T h e Pennsylvania S t a t e University. 1960
(21) J. J o r d a n a n d T. G. Alleman, Anal. Chem., 2 9 , 9 ( l Y i 7 ) .
June 20, 1964 COMPLEX
TRIETHYLAMINE-IODINE 2349

TABLE I
SUhlMARY OF THERMODYNAMIC
DATAFOR 1 : 1 MOLECULAR ADDITIOS COMPOUNDS O F IODISE

B:(ao~v) + Iz(solv)
K,
== +
B : I ~ ( a o i v ) AH,
Lewis base, B: Solvent K , ( t e m p . , " C . ) , 1 /mole - AHr, kcal /mole - A H , , " kcal /mole
N,N-Dimethylpropionamide* CCll 3 9 f 0 . 2 (25) 4 0 f0.3 Kot measurable
N, N-Dimethylaniline' n-Heptane 18 8 ( 2 7 ) 8 . 2 =k 0 3 S o t measurable
S,S-Dirnethyl-p-toluidine' %-Heptane 4 1 . 9 (21 3) 8 3 f0.2 S o t measurable
Pyridined %-Heptane 290 ( 1 6 . 7 ) 7.8 f0.2 8.1 f0 . i
Pyridine' CCI, 97 125) 8 . 0 f 0.5
Triethylamine' n-Heptane 4690 (25) 12 0 11.8f 0.7
Tri-n-propylamine' %-Heptane 1390 (20) 12 1 12.0 i 0.6
Tri-n-butylamine' %-Heptane 1600 (20) 12.3 i 1 . 6 f 0.;
Tri-n-amylamine n-Heptane 12 1 f 0 . 3
Diethylamine' %-Heptane i120 (20) 9.7 8.8 f 0.2
Ethylamine' %-Heptane 720 (20) 7.4 7 5 f 0 6
Determined by thermometric titration; the value of A H r is for the temperature 22 f 2". Deviations are standard deviations.
R . S.Drago, private communication, 1963. e See ref. 2. See ref. 5 . e See ref. 11. f See ref. 4 . y See ref. 3.
TABLE I1 heat evolution was recorded, B-C in Fig. 1. The heat capacity of
EFFECT
O F TRIETHYLAMIXE COXCESTRATIOS O S THE the system a t point C was computed knowing the volume of the
titrant a t this point and assuming the heat capacity varied line-
MAGSITUDE O F A H r MEASUREDI N %HEPTANE
arly with the volume of titrant. The average heat capacity,
Final C,", used in computation was for the region B-C of Fig. 1 . The
C o n c n . of ~ t 3 ~ Ifoles
b
Et3Xa concn., of 1 2 Condition of
heat of reaction, - A h , , was computed from the relationship
titrant, molejl. titrated A H , (mean),C titrated
No mole11 X 10: X 104 kcal /mole solution
I 0 0890 0 810 2 919 11 8 i 0 7 Clear
I1 0.233 2.12 1 . 2 7 1 12 6 f 0 . 4 Clear
where p~ corresponds t o the heat evolved in the formation of 1
111 0.363 3.30 1 . 5 6 8 14 0 f 0 . 4 Clear
mole of product and AT t o the increase in the temperature of the
IV 0 480 4 44 2 154 1 3 . 9 i 0 . 6 Faintly turbid system with the formation of ili moles of product. Because the
1. 0 698 6 35 4 . 0 0 0 14 9 i 0 . 3 Turbid equilibrium constants for 1 : 1 complex formation are sufliciently
VI 0.990 9 00 2.714 1 5 . 1 f 0 . 3 Turbid large and the amines were in large excess, the moles of product
VI1 1.416 12.9 3 . 3 2 1 15 1 f 0 3 Turbid were taken to be equal to the amount of iodine present.
\.I11 0 . 9 3 2 d 8.48" 3 . 7 0 4 12 6 f 0 3 Clear Figure 1 shows typical titration curves for triethylamine-
Concentration of triethylamine titrant before dilution. iodine. The portion B-C is never quite linear even for the most
Concentration of triethylamine titrant after dilution by a fac- concectrated amine titrant solution. The curvature is more
tor of 15.0/165 but before equilibrium was established. This pronounced with the dilute solutions, but the end points remain
is equivalent to the value C,, in eq. 6-8. Deviations ex- remarkably well defined. The excess reagent line C-D is linear.
pressed in terms of standard deviation. Deviations are based on Reference t o Table I shows t h a t the average heats of reaction
four t o five replicates. Total concentration of triethylamine determined by thermometric titration agree well with the values
plus S,S-ditneth~lpropionamide. T h e final triethylamine con- obtained by spectroscopic methods. The values for H , given in
centration was 0.014 M . Table I are for the amine concentrations and iodine concentra-
tions which approximate a s closely as possible those concentra-
tempts t o obtain thermal equilibrium between titrant and titrate tions used in the studies to which comparisons were made.
by thermostatic control of a titrant reservoir external to the These results demonstrate the reliability of the calorimetric
dewar adiabatic titration cell were unsuccessful. By introducing method.
a 17-ml. titrant reservoir filled with titrant into the dewar filled Results
with the solution to be titrated, thermal equilibrium was achieved
within 1 hr. The exact shape of the reservoir depends upon the The dependence of the measured molar heats of
capacity of the dewar and the spatial restrictions within it. Our reaction upon the concentration of the amine in alkyl-
reservoir resembled a miniature cylindrical wash bottle with a amine-iodine systems was investigated in detail for
straight mouthpiece. The reservoir was connected t o a hypo- triethylamine-iodine. The data, summarized in Table
dermic syringe by a short length of Tygon tubing t o form an air-
tight system. For the experiments reported herein the titrant 11, for this system show an increase in the molar heat
was forced from the reservoir a t a constant rate of 7.65 ml. min. -1 of reaction from 11.8 to 13.1 kcal. mole-' as the titrant
as the syringe plunger was moved forward. The delivery rate concentration was increased from 0.089 to 0.991) M .
was faster than that usually employed in order to reduce heat Special care was taken to control the volume of solu-
losses during titration.
Procedure.-A standardized stock solution of iodine in the ap-
tions in these experiments in order to be able to com-
propriate solvent was prepared of such a concentration as to pute amine concentrations a t the termination of a
contain approximately lo-' mole of iodine per 10 ml. of solution. titration
Iodine solutions for use in thermometric titration were prepared
by diluting 10-ml. aliquots of the standard solution to 150 ml. in Discussion
the dewar flask. The amine solutions were prepared by weighing A plot of the average values for the heat of reaction
the amines into volumetric flasks and adding the appropriate and the final triethylamine concentration reveals a
solvent. The titrant reservoir, filled with the amine solution,
was fitted into the dewar filled with titrate and the system allowed near linear relationship a t low amine concentrations.
to come to thermal equilibrium. The heat capacity of the system An explanation for this relationship, which is consistent
was determined in a typical fashion by passing a fixed current for a with all experimental observations, is based on an
measured time through a Manganin coil of known resistance im- equilibrium between Am.12 and Am2.12 complexes ac-
mersed in the titrate. The actual titration was then carried out cording to eq. 2 and 3
followed by remeasurement of the heat capacity of the system.
Because of the significant increase (5-8Yc) in heat capacity of the
system which accompanied the addition of the titrant, i t was Am+ I2 K1= Am.12 + A H i (2)
necessary t o use a value for the heat capacity t h a t represented an K2
average of the system's heat capacity for the period during which Arn,Ip + Am = Am2.12 + A H 2 13)
 8C Ta c t u a l

Rearranging eq. 7 to gi\-e an expression for C2. substi-

tuting into eq. 5, and rearranging gives the equation

6C
-
0
0
c
AH, (h)
._
0-
A s the concentration of the amine increases, the \ d u e

pc0,
C of (K2C.k,,,, - K2C'1?) in the denominator of the second
c
0 term becomes increasingly significant. I t soon pre-
0
40
I actual dominates so that the value of the second term a p
proaches AH2. The expression for AH, in eq. S than
+
L
0
... approaches a constant equal to A H 1 AH2, which is
a
0 equivalent to the heat of reaction of the amine and
I
iodine to form the ;\mz~12complex. Cnfortunately,
in these experiments a t high amine concentrations the
I superposition of heat of crystallization of a solid amine--
20
- 1 iodine complex upon the heat of reaction eliminates thc
possibility o f measuring an accurate value for A H ? .
-At low initial amine concentrations it may be
shown that the curvature of a line connecting plots of
A N , "c,s C.ki,,, is only very slight for values of AFr,
1 I I 1-.
ranging from 2 to 1 kcal. mole--' and values of k'?
0 4 8 12 16 0 4 8 12 It ranging from 20 to 50. (The values for A H ? and K 2are
Volume of T i t r o n t ( m i ) Volume of Titront (mi)
approximate values for the triethylamine -iodine system
Fig. i.---Comparisi:n i)f tlieori.ticnl titrutiim curve. wit11 espcri- obtained from the experimental results.) A theoretical
mental e u r v e i
titration curve was derived by solving eq. 2 and ;3 siiiiul-
The measured heat evoll-ed. q , would be equal to taneously. Reasonable values for the constants K 1 =
47l10 1. mole, K e = 25 1. mole, A H 1 = - 1 1 . 4 kcal.
mole. and A H , = -3.7 kcal. 'mole were assigned.
Theoretical curves are plotted in Fig. 2 for solutions I
where ml, is the number of moles of .Irn.I? I'?7i'fin/iy and V (Table 11). The theoretical and experimental
formed and m2,the number of moles of .lin2.I2 formed curves are in fair agreement in view of the assuniptions
in the titration. The syi~ibolsA H l and AH, are the and approximations t h a t were necessary in rleri\-ing
corresponding molar heats of reaction given in e q . 2 the theoretical curve. At low Concentrations of amine
and #'3. Substituting- conceritraticins for moles in eq. 1 the agreement is quite satisfactory and there is a definite
gives end point in the theoretical curve. The theoretical
treatment makes no allowance for heat developed by
crystal formation nor for heat changes resultirig from a
slight difference in temperature of the titrate and
where 2' is the volunie oi the solution in the dewar titra- titrant.
tion cell. The concentration of the 1 : I complex ill I t is apparent from eq. 8 t h a t when the \ d u e for
oquz'/I'hu'iinz,C'l, i s eyual to the irri'tid concentration oi CAm,is equal to CIi. the second term in ey. S c:mcels.
the 1 : 1 complex. L'l>, minus the coticentration oi the and the heat of reaction is equal to AHl. Despite the
2 : 1 complex that has forinetl. c'.>.and the equilibriurii very slight curvatures at low initial amine concetitra-
arnitie concentration is equal t c J the irii'iini amine con- tion. a linear extrapolation in a AH,z's. c',x,,,, plot would
centratiori. less the conccntratiori o f the 1 : 1 yicld a \-due for AH1 t h a t more nearly represents the
complex initially formcd and the 2 : 1 coniplcs. T h e correct value for this term than the values previously
eyuilibriuiii constaiit for the second equilibrium is rcported for which no compensation was inade for the
therefore presence of the A m 2 . 1 2 species. I t was possible to
ohtairi a value ior AHi of - 11.4 kcal. mole-I for tri-
etliylaniiiie~iodiriein this manner.
T h a t the increased heat of reaction was not due to iii-
creased Coulombic interactions between the highly
The value' ior k',is large. -k(j!)oI . mole and the ainine
- i I polar . h i . I L 'complex and amine molecules (or that this
is present in large excess relative t c i the iodinc so t h a t it contribution was relatively small) was demonstratctl
is a reasonable a ~ ) ~ ~ r ( i x i ~ r to
i ~ ~niakc
t i o t ithe iititial con- by substituting polar N ,S-diniethylpropioiiainide inole-
centration of the I : i complex. (.',,, cqual to the initial cules ( p = :3.W I).) for some of the less polar triethyl-
iotliiie concentratioti. ( ' 1 . Furthcmiiore, the coticeii- aniitic molecules ( p = 0 . 7 5 D . ) . Reference to sample
tratioii of the 2 1 complex is small relati\-e to thc. 1'111 i n Table I shows that the heat of reaction depends
arninc so that - C'i, - L',) is cclu:d apirosi- solely o i i t h e amine concentration and is insensitive to
- c'l,, L-siyg these ai)l)roxitiiations the Imsence of the more polar S ,N-diniethylpropiori-
eq. C i becomes amide in01 ecules.
June 20, 1964
TRIETHYLAMINE-IODINE
Spectroscopic studies of amines and iodine in n-
heptane have established the existence of a 1 : l tri-
ethylamine-iodine complex4 and other 1 1 tertiary
alkylamine-iodine complexes. Direct evidence for
1: 1 complexes of tertiary amines and iodine was pro-
vided by Hassel, who isolated crystals of (CH3)3K.12.18
The isolation of a 2 : 1 triethy1aminee:odine complex
was undertaken in order to provide direct evidence for
its existence. Nagukura reported that for triethyl-
amine-iodine solutions in n-heptane the solutions be-
come turbid and a precipitate readily formed when the
amine concentration was in excess of lo-* M.4 T h e
precipitate was reported “water-soluble. ” We too
have observed such a phenomenon. Solutions of
I? and E t p N in n-heptane, with concentrations compar-
able to samples IV-VI used in thermometric titrations,
developed a turbidity within the time of titrations.
Eventually a brown, gummy material separated from
the solution and the solvent was left colorless. T h e
gummy material solidified upon standing This solid
material was readily soluble in acetonitrile but insolu-
ble in common organic solvents of low dielectric con-
stant. Every attempt to recrystallize this material from
acetonitrile led to the isolation of colorless, needle-like
crystals of triethylammonium iodide, identified by melt-
ing point, elemental analysis, and X-ray spectroscopy.
Despite our repeated failure to isolate the 2 : 1 complex,
the behavior of the product during recrystallization sug-
gested the possibility that an ionic triethylamine-iodine
complex analogous to [ ( y-picoline)J]I’4 is responsible
for these experimental results These iodine(1) com-
pounds are thought to be particularly sensitive to hy-
drolysis and subsequent disproportionation according to
the equation
5[XmzI+][I-] +
3H10 =
2[A4mH-l[Ia-] +
3[AmH*] [ I - ] + [AmH+l [IOS-] + 4Am
(9)
T h e iodine as 1 3 - which is formed in this reaction would
be rapidly and strongly coordinated by the basic amine
which is present in large excess so t h a t some ionic iodine-
(I) species would be regenerated. Hydrolysis and dis-
proportionation of the iodine(1) species would again
set in. This process would continue until all the tri-
iodide is consumed This mechanism is similar to that
proposed by Carlsohn to account for the instability to
moisture of salts containing the dipyridinoiodine(1)
cation.l* Previous success in isolating the 2 : 1 picoline-
iodine complex is thought to be due to the reduced
chemical activity of iodine coordinated to heterocyclic
COMPLEX
2351
amines arising from the distribution of the positive
charge of iodine throughout the aromatic rings. I t
should be pointed out that hydrolysis of iodine(1)
accompanied by disproportionation (eq. 9) would not
change the apparent concentration of titratable iodine.
One equivalent of iodate ion would be generated for
each equivalent of iodine t h a t had disproportionated.
The possibility that the iodine(1) species reacts with the
acetonitrile solvent has not been discounted. A slow
iodination of pyridine is thought to occur when iodine
is dissolved in ~ y r i d i n e . ~ ,h
’ ~reaction of this type would
result in the loss of titratable iodine.
Early workers attributed the time-dependent
change in conductivity of an iodine solution in pyridine
to a slow ionization of a highly polarized species, such
as PYI’+-I”-.~~~ This model has been given a refined
theoretical treatment by Mulliken. 2 2 According to
Mulliken there are two types of charge-transfer com-
plexes, the “outer complex,” B.Xz, and the “inner com-
plex”, [ B X + ][X-1. Mulliken proposed that in the
case of pyridine-iodine, the slow increase in conduct-
ance is due to a slow passage from outer to inner com-
plex over an energy barrier of considerable height.
T h e slow transition from “outer” to ‘‘inner” complex
has been used repeatedly to account for time-dependen t
spectral c h a n g e ~ . ~ J bAn~ ~explanation of the data from
our calorimetric studies requires the existence of two
amine-iodine complexes. But more significantly it
requires t h a t the two species achieve equilibrium in the
time t h a t i t takes to mix the reagents. Experimenta-
tion showed t h a t the measured heats of reaction ob-
tained a t the fastest feed rates were in agreement with
those a t slower rates. From this we concluded that
attainment of equilibrium between the 1: 1 and 2 : 1
complexes is essentially instantaneous. An explanation
for this phenomenon is consistent with the Mulliken
theory if i t is allowed that the energy barrier for the
transition from an outer complex, Am,12, to an inner
complex, [Am21+][I-], is small.
The complexity of amine-iodine systems has been
pointed up by the results of this study. There appears
now to be good evidence in support of the existence of
at least two complex amine-iodine species. Further
investigation is necessary to test the hypothesis t h a t
the interaction of 1 : 1 and 2 : 1 complexes is responsible
for frequently reported time-dependent changes in
conductance and spectra of iodine solutions in nitrogen-
containing basic solvents.
(22) R. S . Mulliken, J . P h y s . Chem., 66, 801 (1952).