lh32
K.
.I. I'ALMtiK
/ C O N T R I R L l I i O N PROM 1 H b G A I bS AND f K I - I LIN
A N D N O R M A N ELLIOI
r
LABORATURILS OP CHhMISTRY NO fi.il 1
Yol. li0 OF THE C A L I F O R N I A I N S T I l U T h 0 1
1 ' 1 I1iNI)I O L Y ,
The Electron Diffraction Investigation of Aluminum Chloride, Bromide, and Iodide BY K. J. PALMER AND NORMAN ELLIOTT 'The unusual physical and chemical properties of aluminum chloride, bromide, and iodide lend considerable interest tc) the electron diffraction investigation of these compounds in the gas phase. Vapor density measurements have shown that ill the gaseous state below approximately 1.00" the substances exist as the dimeric molecules AlZCl6, AlzBrs, and Al&. The same configuration is suggested for these molecules by considerations based on the extreme ionic and the extreme covalent points of view. The radius ratio of the ions Ala-+and C1- is 0.40 (ratio of univalent radiii), which correspoiids to tetrahedral coordination. This can be achieved for a molecule AlzXs by the sharing of an edge between two tetrahedra, as shown in Fig. l . From the covalent point of view this configuration would be expected as the result of the tendency of the aluminum atoms to complete their octet valeiicc shells, the electronic structure of the moleculc being
tion of Fig. 1, with some deformation of the tetrahedra, as described below.
Experimental The electron diffraction photographs were obtained and interpreted in the usual way." The wave length of the electrons was 0.0G13 A. and the camera distance 10.85 cm. for the chloride and bromide and 20.16 cm. for the iodide. The strong tendency of the aluminum halides to hydrolyze made it necessary to transfer the samples to the high temperature nozzle inside a moisture proof box. The x10zzle4 could then be sealed a i d inserted into the electron diffraction apparatus, the sample not being allowed to come into contact with moist air. This procedure proved to be satisfactory, as was verified by inspection of the nozzle after the exposures were made. In no case was there any sign of decomposit io11. Merck c. P. aluminum chloride was used without further purification. The aluminum bro.. mide was made by the method of Richards and '. .AI:'.*AI .. Kre~elka.~ The aluminum iodide was prepared . . .. by heating iodine with excess aluminum in an evacuated glass tube held in a vertical position. 'lhe suggestion of this as a possible structure for The temperature was maintained a t 300" for six the aluminum halides was made by Fajaris.% hours, in which time the color due to the iodine 8 vapor had disappeared completely. The alumimxn iodide which collected in the lower part of the tube along with the excess aluminum was separated from the latter by distilling it to the upper part of the tube and then sealing the tube off a t the center. The product appeared in the iorm of colorless highly refractive crystals. There was no evidence of any iodine vapor being present either during or after the distillation. These crystals 5 were used without further purification. Fig 1 --Thc spatial configuration of Aluminum Chloride.-The photographs of the dimeric molecule A1,XG. Positions 1 and 2 correspond to the aluminum aluminum chloride show nine maxima. The atoms, the remaining positions are ocaveraged values of so, I (the visually estimated incupied by halogen atoms tensities), and C (equal to are given in Table I. The qualitative appearance of the photoWe have carried out the study of these subgraphs is well represented by curve F of Fig. 3 stances by the electron diffraction method, and 1 3 1 1, 0 Brockway, Kea . I f o d o n I'hyr , 8 , 231 (1936) have verified the double-tetrahedral configura1 4 ) I C) Brocks45 and K J Palmer I ' i l I 5 ] O I J H \ A L . 5 9 , 2 1 8 1
:i
G (i
-
51
Sa
A/'>,l
2.24 3.08 4.08 4.9Ii 5.M 6.61 7.43 8.33 9.23 10.11
2.35 2.91 3.93 4.91 5.86 0.51 7.18 8.40 9.32 10.20 10.90 11.91 12.77 13.71 14.6O 15.41 16.30
(1.049) (0.946) ( ,963) ( ,989) 1.000 0.985 ,967 1.008 1.009 1.009 0.99:3 1.00:) 1.005 1.022 1.01Kl 1.013 1.014 I . 008 0.010
10.98
11.90 12.71 8 13.42 S 1 3 14.47 9 15.22 9 2 5 16.07 Average Average deviation Calculated for model F. 7
3
12
TABLETI IN ALUMINUM CHLORIDE INTERATOMIC DISTANCES Kumber of Distance,
2.06 2.21 2.83 3.53 :L56
A.
times distance occurs in molecule
* 0.04
4
f
,IJ4
4
*
.l0
f
.04
1 2
*
;j41*
5.49 * 6.52.t 4.77 *
"OS .%O .l J 3
.05 .15
8
1 2 2
4
Aluslinwn Bromide.-The photographs of aluminum bromide show seven well-defined rings and have the same qualitative features as those for aluminum chloride. The radial distribution curve has two well-defined peaks at 2.28 and 3.7'7 A. The ratio of 3.77 to 2.28 is 1.65, indicating that the tetrahedra in this molecule probably are not distorted to SO great an extent as for the chloride. Curve A of Fig. 4 was calculated for two regular tetrahedra sharing an edge. The curve is in good but not complete qualitative agreement with the photographs. Curves B and C of
Aug., 193s
ELECTRON DIFFRACTION OF ALUMINUM CHLORIDE, BROMIDE, AND IODIDE
Fig. 4 were calculated for models having the same type of distortion as that found for the chloride, but smaller in magnitude. The two models are essentially the same except for the length of the shared edge. In model B this edge was assumed to be 3.36 A. and in model C 3.20 A. The AlzBr3 and A12-Br8 distances were taken equal to 2.21 and 2.33 respectively, in model C, and 2.21 and 2.35 A. in model B. The qualitative agreement of curve C with thc photographs is better than that of curve B in that the fifth maximum in the former is slightly more intense than the fourth, in agreement with the appearance of the photographs. The differences in these two curves are, however, very slight in spite of the fact that the Brb-Br8 distance has been changed by 0.16 The insensitiveness of the intensity curves to variations in this parameter makes it necessary to assign to it a large probable error. In Table I11 there are listed the values of I , C, SO, s (for model C), and the ratio of s/so, and in Table IV there are given the values of the interatomic distances for the molecule and their estimated probable errors.
1855
w.,
I
1
A.
5
10 15 S e. Fig. 4.-Theoretical intensity curves for aluminum bromide.
Aluminum Iodide.-The photographs of aluminum iodide, taken with a camera distance of TABLE 111 20.16 cm., show seven well-defined maxima, the ELECTRON DIFFRACTION DATAFOR ALUMINUM BROMIDE general appearance of the photographs being Max. Min. I C so Sa s ‘sa 1 5 2, 2.11 2.11 1.000 closely similar to that for the chloride and bro2 2.85 2.87 1.007 mide. This similarity is strong evidence for the 2 10 11 3.78 3.72 0.998 assumption that the structures of the three mole3 4.68 4.61 .999 cules are similar in configuration. 6 11 5.48 5.60 1.022 3 The radical distribution curve (curve C, Fig. 4 6.28 6.22 0.990 2), shows principal peaks a t 2.58 and 4.23 8. 4 4 9 7.02 6.88 ,980 5 7.85 7.88 1.004 The ratio of the latter to the former distance is 5 5 13 8.73 8.78 1.006 1.G4, indicating that the structure is very nearly 6 9.59 9.68 1.009 that of two regular iodine tetrahedra sharing an ti 1 2 10.38 10.35 0.997 edge. 7 11.19 11.17 ,998 The ratio of the scattering due to the iodinc 3 11.84 12.08 1.023 7 2 1,003 atoms to t h a t due to the aluminum atoms is very Average Average deviation 0,009 large in aluminum iodide; this makes the determination of the positions of the aluminum atoms a Calculated for model C. with any degree of accuracy impossible. The TABLE IV intensity curves shown in Fig. 5 were calculated INTERATOMIC DISTANCES IN ALUMINUM BROMIDE for models approximating those described for Distance, A. Number aluminum bromide. Curve A is for undistorted A12-Br1 2.21 * 0.04 4 tetrahedra, and curves B and C for tetrahedra AkBra 2.33 * .04 4 whose shared edge has the value 4.00 and 3.85 k . , Brb-Bra 3 . 2 0 * .10 1 Br3-Br4 3 . 7 2 * .03 2 respectively, and for which the A12-13 and A12-Is Bra-Brs 3.78 * .03 8 distances have the values 2.58 and 2.54 k . , reA4-AL 3.39 * .10 1 spectively. Curve A does not agree with the Brs-Brr 5.76 * .10 2 photographs in that the relative intensities of the B~e-Bre 6.86 * .10 2 maxima are unsatisfactory. Curve C gives a -11 - E r , 4.93 * .10 4
somewhat better represeiitatioii o l the appearance of the photographs than cur\-c 13, hut the differenccs in these two curves, namely, the variation of the iiiteiisities of the third, fourth, and fifth maxima, are so small that, as in the case of the bromide, i t is impossible to determine thc length of the shared c d p with miic.h accuracy.
I'ABlt.
\'I
IKThRATOhUC 1)ISTANCES I N ALUMINUM IODIDE
1)rstancc
AI, I, I, In I Q JI 11 I7 I, AI) ,111 1 1A12
I
A
3 53 = 0 04 2 58 * 04 2 90 L 1.5 1 211 + 0; t 21 t 03 ,;:'ti 15 1121 i i 3 i 82
16
-11- ii,
+ 3-
i-
1.5 LO 1.5
Xumber
4
4 1 2 8
1 d
2 1
Discussion
10
-I
5 -. Fig. 5.--Theoretical intensity cilrves for aluiuinum iodide.
The values of the interatomic distances with their estimated probable errors are given in Table VI. The quantitative comparison of the s values obtained from curve C with the observed SOvalues is given in Table V. TABLE
1'
ELECTRON DIFFRACTION DATAFOR ALUMINUMIODIDE Max
Min.
1
I
C
so
$16
9
3
1 85 2 51
1 87 2 53
10
9
3 33
7
IU
3 32 4 14 4 85
5 55 9 t i 20 7 03 10 7 80 8 62 3 9 27 9 95 1 10 67 Average
.j 65
2 2
3 3 4
6
4 6
5
5 6 2
6
I
7
1
4 08 4 YO 0 313 7 00 7 75 8 67 9 29
9 89 10 70
Average deviation
Calculated for model C
J
so
0 995
I 1 0 1 1
004 00.3 981;
029 0 lti I)9 9 j 996 994 1 OOti 1 002 0 994 1 003 1 oo:! 0 008
'I'he oiily report on the structures of aluminum chloride, bromide, or iodide previous to the pres( s i l t one is that of Ketelaar' on the structure of aluminum chloride crystals. He found that the chlorine atoms are in hexagonal closest packing, this arrangement being compatible with that found for the gas molecule in this investigation. However, he chose to place two aluminum atoms inside an octahedron of chlorine atoms, and only 0 3fi A. apart, rather than one each inside of two tetrahedra sharing an edge, both of these possibilities being provided by the hexagonal closest packed arratigement. The extent to which the X-ray data can be accounted for by this latter configuration is being investigated by one of us. Curve G, Fig. 2 , is the .simplified theoretical intensity curve calculated for the "octahedral' inodel of Ketelaar; it is apparent from a comparison with curve F that this model cannot represent the structure of the gas molecule. 'I'hc large difference in electronegativity between aluminum and the halogen atoms leads one t o expect that the -41-X bond will be largely ionic, and this is confirmed by the observed contraction of the shared edge. The percentage decrease i n length of the shared edge is found to be largest in the chloride and least in the iodide, which is 111 accordance with expectation. The sums of the tetrahedral radius of aluminum and the normal radii for the halogen atoms are 2 25, 2.40, and 2 59 for the chloride, bromide, and iodide, respectively. These values are to be compared with the observed values, 2.06, 2 21, and 2.53 A.,which are the A12-Xa distances, and 2 3 1, 2 33, and 2.5h which are the A12-Xs distances, for the chloride, bromide, and iodide, respectively. The observed shortening in the case o f the Alq-X3distalices is probably due to the exvitctl structures in which an X? halogen atom
a.,
"I
1
\
?i K e t i l r a i / hi c i
90,217il'lRil
THEF L U O R E S C E N C E
Aug., 1938
swings in a pair of electrons and forms a double bond with the aluminum atom. This type of resonance is not expected to occur to the same degree for halogen atoms forming two bonds, which accounts for the fact that the observed A12-Xs distances are nearly equal to the sum of the appropriate radii. I t is interesting to note that the observed values of the A12-X3 distances show a greater tendency for the chlorine and bromine atoms to form double bonds than of iodine atoms; this is compatible with the results of other investigations. We wish to express our thanks to Professor Linus Pauling for his aid and helpful criticism during the course of this investigation.
Summary It is shown that in the gaseous state the dimeric
[ C O N T R I B U T l O N FROM T H E
OF
1857
ACETONEV A P O R
molecules of aluminum chloride, bromide, and iodide consist of two tetrahedra sharing an edge with six halogen atoms a t the corners, each tetrahedron containing one aluminum atom. The final values of the interatomic distances are as follows The subscripts on the atomic symbols refer to their positions in the molecule as given in Fig. t . ~
1
~
A11-A12 3.41 * 0.20 A. A12-x~ 2 . 0 6 * .04 A12-Xs 2.21 * .04 AlZ-Xe 4 . 7 7 * .15 Xs-Xd 3 . 5 3 * .04 x3-X~ 3 . 5 6 * .02 x 6 - X ~ 2.83 * .10 XB-X~5.49 * .05 X r X s 6 . 5 2 * .05
1
AIzBre ~
~
1
~
1
3.39 * 0.10 ii.3.24 * 0 . 1 5 ii. 2.53 * .04 2 . 2 1 * .04 2.58 * .04 2.33 * .04 5.22 * .15 4.93 * .10 4.20 * .03 3 . 7 2 * .03 4.24 * .02 3 . 7 8 * .03 2 . 9 0 * .15 3 . 2 0 * .10 5 . 7 6 * .10 6 . 2 4 * .15 6.86 * .10 7 . 5 4 * . l o
PASADENA, CALIF.
RECEIVED MAY31, 1938
METCALFC H E M I C A L LABORATORY O F BROWNU N I V E R S I T Y ]
Photochemical Studies. XXVI. A Further Study of the Fluorescence of Acetone Vapor and its Relationship to the Photochemical Decomposition BY MAXS.MATHESON AND W. ALBERTNOYES,JR. The fluorescence of acetone vapor was reported by Damon and Daniels,' who stated that it is greenish and is changed to blue by the presence of small amounts of oxygen. The oxygen disappears during illumination and this fact has been used as a means of determining small amounts of this substance.2 Two regions, one extending from 4100 to 4820 A. (maximum 4580 A.) and the other from 4990 to the limit of plate sensitivity (5210 A.), were found by Damon and Daniels.' Norrish, Crone and Saltmarsh3report three bands of fluorescent emission (all diffuse) with maxima a t 5117, 5572 and 6095 A. More recently Padmanabhan4 has found that the bands are not devoid of structure but consist of diffuse bands superimposed upon a continuous spectrum. Ten bands, two considered to be doubtful, were reported. Herzberg6 first found a discrete structure in the near ultraviolet absorption of acetone vapor and (1) Damon and Daniels, THISJOURNAL, 55, 2363 (1933). (2) Damon, I n d . Eng. C h e m . , A n a l E d . , 7 , 133 (1936); Fugassi, THISJOURNAL, 59, 2092 (1937); Fugassi and Daniels. i b i d . . 60, 771 (1938). (3) Norrish, Crone and Saltmarsh, J . Chem. Sac.. 1456 (3934). (4) Padmanabhan, P r o c . I n d . A c a d . Sci., SA,,594 (1937). (5) For mention see Scheibe. Povenz and LirtsttBrn, Z. p h y s i k . C h e m . , 4 0 8 , 302 (1933).
The this has been confirmed by other detailed analysis of these bands and their interpretation is still lacking, but the existence of such a structure is in conformity with the fact that fluorescence is observed. However, structure is difficult to observe in many polyatomic molecules even though fluorescence is found. Such seems to be the case with ethyl methyl ketone7 and diethyl ketone.* All authors who have studied the quantum yield of acetone decomposition in the near ultraviolet a t room temperature agree that it is The explanation of this fact is not complete. Damon and Daniels' state that the fluorescence is too weak to be the primary cause of this effect. However Fisk and Noyeslo studied the fluorescence excited by the 3130 A.line of mercury and found (with the very low intensities used) a Stern-Volmer mechanism to be obeyed. At moderate pressures it was found that the number of quanta absorbed was roughly equal to the sum (6) Noyes, Duncan and Manning, J . Chem. P h y s . , a, 717 (1934); Noyes, T r a n s . F a r a d a y Sac., 33, 1495 (1937). (7) Duncan, Ellsand Xioyes, THISJOURNAL, SS, 1454 (1936). (8) J . W . Zabor, M.S. Thesis, Brown University, 1938. (9) See refs. 1 and 3, L.eermakers, THIS JOURNAL, 66, 1900 (1934). (10) Fisk and Noyea, J . Chem. P h y s . , $4, 654 (1934).
~
