close

Вход

Забыли?

вход по аккаунту

?

Studies in the chemistry of sulfur nitride

код для вставкиСкачать
STUDIES IN THE CHEMISTRY OF SULFUR NITRIDE
A Dissertation
Presented to
the Faculty of the Department of Chemistry
The University of Southern California
In Partial Fulfillment
of the Requirements for the Degree
Doctor of Philosophy
by
Don L. Armstrong
August 1941
UMI Number: DP21730
All rights reserved
IN FO R M A TIO N TO ALL U S E R S
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
Dissertation Publishing
UMI D P 21730
Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 - 1 3 4 6
P/lv
^
^
£/
XTtzj d is s e r t a t io n , w r i t t e n b y
...L«...MMSTROM..........
u n d e r th e g u i d a n c e o f h.%Q F a c u l t y C o m m i t t e e
on S t u d ie s , a n d a p p r o v e d b y a l l its m e m b e rs , has
been p r e s e n t e d to a n d a c c e p te d b y th e C o u n c i l
on G r a d u a t e S t u d y a n d R e s e a rc h , in p a r t i a l f u l ­
f i l l m e n t o f r e q u ir e m e n t s f o r th e d e g re e o f
D O C T O R O F P H IL O S O P H Y
D ean
Secretary
Dfl*e...Septgmbe.r..X5.,... 1941
Com m ittee on Studies
Chairm an
&
e.
L
f.
J
L
/ «
l
,/
TO MY WIFE
TABLE OF CONTENTS
CHAPTER
I.
II.
III.
IV.
PAGE
INTRODUCTION..................................
1
Introduction to the p r o b l e m ................
1
Statement of the problem....................
3
REACTIONS OF SULFUR N I T R I D E ..................
4
Hydrolysis .................................
4
Oxidation
and instability
5
Reactions
of preparative interest
Reactions
of structural importance .........
8
PREPARATION
OF SULFUR N I T R I D E ...............
12
Review of
previous methods .................
12
. . . . .........
. . . . . .
CONSTITUTION OF SULFUR NI T RI DE ............
6
16
Comparison and criticism of suggested struc­
tures.
.................................
Structure suggested by this w o r k ............
V.
EXPERIMENTAL
................................
Preparation of sulfur nitride
..............
16
21
23
23
Purification by vacuum-sublimation . . . . . .
26
Analysis of the p r o d u c t .............
28
Preparation of the boron halides ............
29
Apparatus and technical methods
............
30
Action of electron acceptors ................
34
Preliminary experiments
.
..............
34
Experiments with the final apparatus . . . .
36
iv
CHAPTER
PAGE
Stability of the addition compounds ..........
42
71. DISCUSSION........................................ 44
Discussion of experimental results
711.
..........
S U M M A R Y ........ •....................
BIBLIOGRAPHY...............
.
44
47
49
LIST OF TABLES
TABLE
I.
II.
PAGE
Boron Fluoride— Sulfur Nitride
Ratio
Boron Chloride— Sulfur Nitride Ratio
. . . . . . . 44
..........
45
LIST OF FIGURES
FIGURE
PAGE
1.
Diagram of Preparation Assembly...............
25
2.
Diagram of High Vacuum A p p a r a t u s .............
32
3.
Diagram of All-glass, Magnetic V a l v e .........
33
CHAPTER I
INTRODUCTION
I.
INTRODUCTION TO THE PROBLEM
From the moment of its discovery, over one hundred
years ago, sulfur nitride has been the source of many contra­
dictions and of much uncertainty.
This doubt has extended
even to the determination of the molecular weight of the
compound.
In view of carefully done later work, the weight
of one mole has been fixed at 184.
A combination of this
information with an elemental analysis leads to the accepted
formula, S^N^.
In spite of the relatively small number of
atoms per molecule, the structure is still being debated,
with little hope of an early decision.
The determination of the above formula was not accom­
plished until nearly fifty years after the discovery of the
1
compound.
Although Soubeiran
was the first to analyze sul­
fur nitride, the method of purification which he used did not
completely separate the sulfur nitride from the free sulfur.
His analysis showed the presence of 77 per cent sulfur and
22 per cent nitrogen.
The theoretical values are 69.56
per cent sulfur and 30.44 per cent nitrogen.
Fordos and Gells
^ Soubeiran, Ann. Chim. Phys. 67. 71 and 96 (1938).
2
Fordos and Gelis, Comp, rend. 31. 702 (1850).
2
obtained reasonably accurate results by the use of carbon
disulfide as an extractive agent.
However, since the atomic
weight of sulfur was assumed to be 16 at that time, they coneluded that the simplest formula was NS 2 . It was Schenck
L
who, according to Lengfeld and Stieglitz, determined the
3
molecular weight by the cryoscopic method with naphthalene as
a solvent.
His value of 184 corresponds to a molecular
formula of S^N^.
Although it was known that carbon disulfide
attacks sulfur nitride at room temperatures, Muthmann and
5
Clever *used this solvent to determine the molecular weight
of sulfur nitride by measuring the molecular elevation of the
boiling point.
Perhaps fortuitously, they reported a value
which agreed with that of Schenck.
Francis and Davis
7
6
With greater certainty,
were able to check this figure by a similar
measurement using benzene as the solvent; hence it has been
definitely accepted as correct.
No such agreement is found in the literature concerning
the crystal structure.
Sulfur nitride is an orange-colored
3
R. Schenck, Naturforschenden Gesellschaft zu Halle
a. S. February (1895).
^ F. Lengfeld and J. Stieglitz. Berichte 28. 2742
(1895).
'
5
A. Clever and W. Muthmann, Berichte 29. 340 (I896 ).
6
Loc. cit.
^ F. E. Francis and 0. C. M. Davis, £. Chem. Soc. 85,
259 (1904).
3
crystalline material, having reported decomposition points
ranging from 159° C to 180° C.
Zanstra
It was reported by Jaeger and
ft
as belonging to the orthorhombic system of crystal9
line structure; however, Buerger states definitely that this
opinion is in error, and that the substance should be classi­
fied as monoclinic.
However, all investigators agree that
there are four molecules in the unit cell.
II.
STATEMENT OP THE PROBLEM
Information concerning the structure of sulfur nitride
has been obtained by a consideration of the reactions which
are peculiar to this compound.
Some of the reaction products
have been in the nature of complex addition compounds, but
there is a great degree of uncertainty in the interpretation
of knowledge concerning these.
It was thought that informa­
tion leading toward the solution of this problem could be
obtained through the study of relatively simple addition com­
pounds.
Plans for these compounds were made in such a manner
that there could be no doubt regarding the nature of the bond
between sulfur nitride and the additor compound.
Several
electron pair acceptors of varying strength were allowed to
react with sulfur nitride in such a manner that the effect of
each could be quantitatively determined.
g
F. Jaeger and J. Zanstra. Proc. Acad. Sci. 3A.
782 (1931), Chem. Abs. 26, 2098 (193^T7
^ M. J. Buerger, Am. Mineral. 21. 575 (1936).
CHAPTER II
REACTIONS OF SULFUR NITRIDE
I.
HYDROLYSIS
The action of water upon sulfur nitride, either in
acid, neutral or basic media, always was thought to produce
ammonia or its salts quantitatively.*^
According to Meuwsen,”**
the equation:
S4N4 + 6NaOH - 2Na2S03
Na2S2C>3 ♦ 4NH3
represents the hydrolysis in a basic solution.
Since the
process is gentle, it is concluded that the molecule contains
no nitrogen to nitrogen links.
Furthermore, since no hydrogen
sulfide is produced, the compound is properly to be considered
12
a nitride rather than a sulfide.
The production of thiosulfate ion has been taken as evidence of at least one sulfur
to sulfur linkage per molecule.
Serious objections can be offered to the above inter­
pretations of the hydrolysis reaction.
The absence of hydro­
gen sulfide can be accounted for by recalling that alkali
hydrogen sulfides and hydrogen sulfides interact to produce
Fordos and Gelis, oj). cit.. 702;
R. Schenck, Ann, der Chimie. 290. 171 (1896 ).
^ A. Meuwsen, Berichte 62B. 1959 (1929).
12
M. Arnold, J. Hugill, and J. Hutson, J. Chem. Soc..
1645 (1936).
”
5
thiosulfates.
Free sulfur also reacts with sulfites to yield
thiosulfates.
It is known that free sulfur is produced dur­
ing the hydrolysis of sulfur nitride in acid solution and it
is an open question whether this sulfur is formed directly or
by the decomposition of thiosulfate ions.
Therefore, one can­
not conclude that the sulfur to sulfur bonds existed in the
original compound, merely because thiosulfates are produced.
A further bit of information concerning the state of
the sulfur atoms in the compound was obtained by Ruff and
13
Geisel
who expressed their quantitative results by the
equation: S^N^ + 12HC1 = 4NH3 + 4(S+3C1).
They wrote the
equation in this manner in order to emphasize their deduction
that the four sulfur atoms account for a total of twelve
valence bonds.
Meuwsen^ attacked this problem from an en­
tirely different viewpoint.
He discovered that, "tetrahydro
sulfur nitride,” (tetrathiazol) required 16 equivalents of
bromine to oxidize the sulfur to sulfate.
Assuming that the
hydrogen was linked to the sulfur atoms, he concluded that
each of the latter had a valence of 4 .
II.
OXIDATION AND INSTABILITY
The air-oxidation of sulfur nitride often occurs
13 0. Ruff and E. Geisel, Berichte 37. 1573 (1904).
Meuwsen, op>. cit.. 1959.
6
explosively, especially if the temperature is near the point
at which decomposition occurs (165° C).
Several observers
warn against the instability of sulfur nitride in air, even
at room temperatures, saying that it is subject to spontane­
ous decomposition.
It must be noted, however, that persons
issuing such warnings used carbon disulfide in the extractive
process.
Apparently if this solvent is avoided, as was done ~
throughout this study, the danger of an explosion is lessened,
if not completely avoided.
The action of concentrated acids is extremely vigorous
if surface wetting occurs.
In one analytical process, boiling
aqua regia was ineffective as an oxidizing agent because poor
contact was made.
On the other hand, sulfuric acid digestions
could not be used in determining nitrogen, as mild explosions
occurred.
III.
REACTIONS OF PREPARATIVE INTEREST
Numerous other reactions are of interest in relation
to the preparation of sulfur nitride.
The action of halogens
on sulfur nitride is important since free chlorine is present
in small quantities during the reaction between ammonia and
sulfur dichloride, due to the decomposition of the latter.
Meuwsen1^ reported the formation of compounds having the
15 A. Meuwsen, Berichte 6AB. 2311 (1931).
7
formulae (NS^Cl and {NSC1)3 as a result of the treatment of
sulfur nitride with chlorine.
Voznesenskii
16
reported also
a simple addition of four atoms of chlorine to one mole of
sulfur nitride.
Using a carbon disulfide solution of sulfur
17
nitride, Muthmann and Clever
obtained tetrabrominated sul­
fur nitride and by allowing bromine vapor to attack solid
sulfur nitride, produced a hexabromo compound.
The presence of free chlorine, due to the decomposition
of sulfur dichloride, indicates that sulfur monochloride is
18
also present during the preparation reaction. DeMarcay
and
19
WSlbling
reported the production of thiotrithiazyl chloride
(S4N 3CI) from the treatment of sulfur nitride with sulfur
20
monochloride. DeMarcay
also indicated that he found an ad­
dition compound, which consisted of/one mole of sulfur nitride
in combination with one mole of sulfur monochloride. A simi21
lar addition compound was obtained by Voznesenskii
by the
use of sulfur dichloride in place of the monochloride.
S. Voznesenskii. J. Russ. Phys* Chem. Soc. 59.
221 (1927).
17
18
Clever and Muthmann, op. cit., 340.
E. DeMarcay, Comp, rend. 91. 1066 (1880).
19 H. WSlbling, Z. anorg. Chem.
20
21
DeMarcay, op. cit.. 854.
Voznesenskii, op. cit.. 221.
2L>
281 (1908).
The
8
same author reported a reaction between sulfur nitride and
22
ammonia to form a diammoniate, while Meuwsen
stated that
23
there also exists a triammoniate. Usher
asserted that under
certain conditions, even sulfur reacted with sulfur nitride
to produce a compound containing two sulfur atoms per nitro­
gen atom.
From the preceding consideration of the substances
present during the formation reaction, one is led to certain
conclusions regarding the conditions which must prevail if
good yields are to be obtained.
It is believed that these
conditions were obtained by conducting the preparation in a
manner which will be described in detail in a later section.
IV.
REACTIONS OF STRUCTURAL IMPORTANCE
Numerous reactions of sulfur nitride are more interest­
ing from the structural than the preparative viewpoint.
24
Voznesenskii
reported that he obtained the previously men­
tioned thiotrithiazyl chloride from the treatment of sulfur
nitride with acetyl chloride.
this result.
No explanation was offered for
In liquid ammonia solutions, sulfur nitride is
attacked by lead iodide and mercuric iodide, yielding
22
23
Meuwsen, Berichte. 62B. 1959 (1929).
F. Usher, Journ. Chem. Soc. 127. 730 (1925).
Loc. cit.
9
substances having the formulae PbN2S 2 and HgJ^S respectively.^-5
26
These substances have been explained by Ruff and Geisel
on
the basis of a certain assumed structural formula for sulfur
nitride.
The structures thus derived are as follows:
Pb
S:SSK
/ \
N sS: N
S
S :N
I
=
N
I
and
,hn
N : Hg = SL.
Hg
Pb
These authors preferred to consider that the metallic atom
in the compounds was joined either to one or to both of the
nitrogen atoms.
They assumed the existence of two isomeric
forms of the reaction products as shown. Such a picture is
27
contrary to the belief of Meuwsen
who stated that the lead
and mercury are bonded to the sulfur atoms.
Both views re­
quire the existence of a sulfur to sulfur linkage in the
original nitride since such a bond would be unlikely to form
under the conditions of the reaction.
In direct contrast to the above assumptions is the
28
work of Schenck.
In his study of the problem, he found
that electron donors of the tertiary amine type did not react
with sulfur nitride nor did secondary aromatic amines.
25
Ruff and Geisel, op. cit.. 1573;
Meuwsen, Berichte. 64-B. 2311.
26
Loc. cit.
27
28
Loc. cit.
Schenck, o£. cit.. 171.
However,
10
when secondary aliphatic amines were allowed to react with
sulfur nitride, substances of the type I^NSNRgwere obtained
quantitatively.
This was later refuted by Ruff and Geisel;
29
however, if the reaction were quantitative it would indicate
that no sulfur to sulfur linkage exists in sulfur nitride.
Electron acceptors react quite differently.
The most
characteristic action of an electron acceptor is the forma­
tion of an addition compound in the ratio of one mole of additor to one mole of sulfur nitride.
An example of this is
the case of sulfur chloride previously mentioned.
Such com­
pounds are also formed with stannic chloride, antimony pentachloride, and tungsten p e n t a c h l o r i d e T h e r e is a tendency
to overestimate the importance of these reactions.
It must
be noted that because of the large size of the central atom,
it is possible that double attachments of sulfur or nitrogen
to these polychlorides may occur.
If sulfur nitride thus
might have chelate behavior, such compounds are of little
benefit in furnishing information regarding the structure.
The use of electron acceptors having a maximum coordination
number of A avoids bridged polymeric addition compounds.
the boron halides fall within this class, they were used
throughout this work.
Ruff and Geisel, oj>. cit.. 1573.
DeMarcay, o
0. Davis, !T7 Chem. Soo. 89. 1575 (1906).
WSlbling, op. cit.. 281.
As
CHAPTER III
PREPARATION OF SULFUR NITRIDE
I . REVIEW OF PREVIOUS METHODS
The literature bearing upon the preparation of sulfur
nitride is highly contradictory and difficult to interpret.
The most common method depends upon the reaction of ammonia
with one or more of the sulfur chlorides, a process by which
31
the original discoverer, Gregory,
obtained his evidently
impure samples.
As developed by him, the procedure followed
was to add, "liquid" ammonia dropwise to sulfur chloride.
Since ammonia had not been liquified at the time this work
was done, it can only be assumed that what Gregory termed,
"'liquid ammonia," was in truth an aqueous solution of the gas.
The presence of water during the reaction has been decried by
some later workers on the ground that the yield is materially
32
reduced. Soubeiran
was the first to recognize this fact
and avoided the presence of water by drying the ammonia over
caustic potash.
His method of extraction evidently was not
very successful, since an analysis revealed the presence of
33
excess sulfur. Fordos and Gelis
were the first to use the
(1836 ).
32
33
Gregory, Journ. Pharm. 21, 315 (1835), 22, 301
“
“
Soubeiran, op,, cit.. 71 and 96.
Fordos and Gelis, 0£. cit.. 702.
12
higher chlorides of sulfur which they termed, "sulfur perchloride."
They used carbon disulfide as an extractive, al­
though it was realized that sulfur nitride was attacked by
the solvent.
Heretofore, all solvents for the reaction mixture had
been avoided.
The use of benzene as a solvent for the sulfur
3/
35
dichloride was introduced by Schenck.
Ruff and Geisel
used benzene as a solvent and, "sulfur chloride," containing
69 per cent chlorine and were the first to suggest an equation
for the reaction between ammonia and sulfur dichloride:
16 NH 3 + 6SC12 * S
♦ 2S + 12 NH^Cl.
Basing their calculations on this equation, they re36
ported yields as high as 50 per cent. Meuwsen
developed
this technique considerably, paying particular attention to
the elimination of water from every possible source; however,
this apparent improvement of method did not noticeably in37
crease the yield obtained. Van Valkenberg and Bailar
sub­
stituted diethyl ether for benzene as a solvent.
They re­
ported yields up to 65 per cent using either sulfur mono­
chloride or sulfur dichloride.
3L
After a succession of poor
Schenck, 0 £. cit.. 171.
^ Ruff and Geisel, op_. cit.. 1573.
36
Meuwsen, Berichte 65B. 1724 (1932).
37
H.
Van Valkenberg and J. Bailar, J. Am. Chem. Soc
47, 2134 (1925).
13
results, Arnold, Hugill and Hutson
solvent is chloroform.
38
concluded that the proper
Realizing that dilution is important
in reducing the effect of side reactions, they dissolved sul­
fur monochloride in chloroform in the ratio of one to ten.
They reached the rather startling conclusion that the presence
of small amounts of water is unimportant.
The most recent
report of the preparation has been contributed by Swinehart.
39
His method does away with a liquid solvent and substitutes an
inert solid such as ammonium chloride or bentonite clay upon
which the sulfur dichloride is absorbed.
Gaseous ammonia is
passed over the solid with appropriate control of temperature
and pressure.
It will be noted that all of the methods pre­
viously mentioned have one detail in common, namely that the
sulfur chloride is always present in excess until the reaction
is completed.
Some authors mentioned the use of a minimum
quantity of ammonia, while others continue its addition for
sometime after the main reaction is ended.
numerous reactions which can
an excess
occur
of either reagent, it is
In view of the
between sulfur nitrideand
surprising that suchhigh
yields have been reported.
Ruff and Geisel^0 extended their work to an entirely
38
Arnold, Hugill and
Hutson, op. cit.. 1645.
39
Carl F. Swinehart, U. S. Patent 2,190,177,
February 13, 1940.
Ruff and Geisel, Berichte 38. 2659 (1905).
u
new method of preparation.
This process is based on the
highly reversible reaction between sulfur and liquid ammonia,
which is supposed to produce ammonium sulfide and sulfur
nitride.
The equilibrium is displaced in the desired direc­
tion by the addition of silver iodide which, itself soluble
in liquid ammonia, serves to precipitate the sulfide.
41
Bergstrom
was able to show that the simple equation of Ruff
42
and Geisel:
10 S + 4 NH3 - 6H2S + S4N4
does not represent the entire truth of the process.
He main­
tained that under such conditions, sulfur nitride reacts ^‘
d.th
the excess sulfur present and further that ammonium polysul­
fide is produced.
Each such side reaction vrould, of course,
decrease the yield obtainable.
43
Moldenhauer and Zimmerman
produced sulfur nitride by
direct combination of the elements under the influence of an
electrical discharge at 12mm pressure of nitrogen.
Since the
fraction of the nitrogen activated under such conditions is
very low, side reactions of even small magnitudes seriously
affect the yield.
(1929).
This method like the more common one, yields
^
F. Bergstrom, J. Am. Chem. Soc. 48. 2319 (1926).
4-2
T
'4 4.
Loc.
cit.
^
W. Moldenhauer and A. Zimmerman, Berichte 62B. 2390
15
nitrides of sulfur other than the one desired.
Attempts of the author to repeat the work of Meuwsen^
45
46
of Ruff and Geisel,
and of Moldenhauer and Zimmerman
led
to no evident success.
This lack of laboratory reality, as
well as the obvious confusion in the literature, made it nec­
essary to develop a new and dependable technique for the
preparation of sulfur nitride.
By the method finally adopted,
the destructive side reactions are minimized by conducting the
ammonolysis in the gas phase and avoiding an excess of either
reactant.
The process is described in detail in the experi­
mental section.
****
45
46
Meuwsen, Berichte, 65B. 1724.
Ruff and Geisel, Berichte 38. 2659.
Moldenhauer and Zimmerman, op. cit.. 2390 .
CHAPTER IV
CONSTITUTION OF SULFUR NITRIDE
I.
COMPARISON AND CRITICISM OF SUGGESTED
STRUCTURES
One of the earliest structures for sulfur nitride was
47
suggested by Schenck:
N
/\
\/
S'S
— ■ N
/\
s
s
\ /
N -■” ■== N
His reason for proposing such a structure was to avoid a sul­
fur to sulfur bond, (which he concluded was absent, as a re­
sult of his work with secondary aliphatic amines). Ruff and
48
Geisel
criticized his conclusions severely, since his own
report mentioned the lack of production of hydrazine upon
complete reduction of sulfur nitride.
A residue containing
nitrogen to nitrogen links of some sort might well have been
expected from such a structure as he suggested.
49
Arnold, Hugill, and Hutson,
in discussing their own
views on structure, credited the following structure to
47
R. Schenck, Naturforschenden Gesellschaft zu Halle
a.S. February (1895).
48
Ruff and Geisell, Berichte X L , 1573 (1904).
^
Arnold, Hugill, and Hutson, oj>. cit., 1645.
Muthmann and Clever,
50
however, a very careful search of the
paper referred to, as well as another by these authors,
failed to reveal any suggested structure.
One of the structures proposed by Huff and Geisel
has persisted in favor since its origin and even appears in
a modernized form as one of the three phases of the resonance
52
system suggested by Arnold, Hugill, and Hutson:
N:§
S :N
N=S=N
Their reasoning was as follows.
of sulfur in the compound is three.
The average valence
Since this value is not
known in other sulfur compounds, it is reasonable to assume
that sulfur is bonded to sulfur, in which case all the sulfur
atoms would be tetravalent or a condition of mixed valency
exists in the divalent, tetravalent and possibly hexavalent
forms.
Assuming that the first proposition represents the
truth, another structure must be presented, namely:
/S = N -S
I
'^3— N-S7'
n
50
Clever and Muthmann, op,, cit.. 340.
^ Huff and Geisel, Berichte 37. 1573.
52
.
,
Arnold, Hugill, and Hutson, op,, cit.. 1645
If, on the other hand, a condition of mixed valency exists,
several other structures are possible, among which is the
53
first. Repetition by Ruff and Geisel
of the work of
Schenck,^ led to only a 50 per cent yield of thiodipiperidine.
Inasmuch as this reaction is not quantitative, it is
quite possible to have sulfur to sulfur bonds.
The action
of lead iodide on sulfur nitride in liquid ammonia does not
eliminate from consideration the two formulae offered by
Ruff and Geisel.
The action of ammonia on sulfur nitride is
analogous to the action of water in the production of sulfurous and thiosulfuric acids.
The mechanism for the action of
lead and mercuric iodides is now easily explained on the basis
of the first formula thus:
N:S
S:N
ti-g-Hf"'
+
2N%
=•
N:S
2 N:H2 +
H-N=S=N-H
then
§
(II)
(I)
(I) + Hg=|:N
and (II)
+
Pb = .Pb
Hg
Ruff and Geisel did not choose between the suggested
structures, but merely indicated their preference for the
55
56
first. Meuwsen
used the report of Ruff and Geisel
on the
^
Ruff and Geisel, Berichte 37* 1573.
^ R. Schenck, Ann, der Chimie. 290. 171 (1896)
55
Meuwsen, Berichte 62B, 1959.
56
Ruff and Geisel, Berichte 37. 1573.
19
action of hydrogen chloride with sulfur nitride as evidence
that their first formula was incorrect, since the four sulfur
atoms contain a total of twelve val-ence bonds.
As previously
stated, Meuwsen worked toward the structure of sulfur nitride
from a study of tetrathiazol which was first prepared by
57
WSlbling.
He considered that the thiazyl group (SN) is
analogous to the cyanide radical (CN) as sulfur dioxide is to
carbon monoxide and therefore, he concluded, that sulfur ni­
tride is analogous to dicyanogen (CN)2 .
Inasmuch as tetra­
thiazol reacts with formaldehyde to produce (NSCH20H)^,
58
Meuwsen
considered that this action was evidence for the
conclusion that the hydrogen atoms were bonded to the sulfur
and not the nitrogen, since if the latter were true, removal
of water by formaldehyde should yield (SNCH2NS}2 . He con­
sidered the former analogous to that action which occurs be­
^
tween formaldehyde and sodium bisulfite. On the basis of
these data and assumptions, Meuwsen 59 concluded that tetra­
thiazol is the nitrile of the unknown acid-HS02H.
Similarly,
sulfur nitride was thought to be the cyclic nitrile of two
moles of disulfinic acid (I^SgO^.).
He presented the following
structure for tetrathiazol and endorsed the second presented
N"'
^ WSbling, o£. cit,, 281,
58
Meuwsen, Berichte 62B, 1959.
59 Loc. cit.
by Ruff and Geisel for sulfur nitride.
Working from this structure, he proceeded to offer explanations for several of the reactions of sulfur nitride.
60
Some of the postulated intermediate steps for some of the
reactions seem rather illogical; as for example, the action
of ethyl magnesium bromide upon sulfur nitride was pictured
as producing an intermediate addition compound which broke up,
61
"because the molecule was too large."
The modernization of the first formula presented by
Ruff and Geisel was accomplished by Arnold, Hugill, and
62
Hutson
by including it as one phase of a resonance system.
They objected to the others proposed on the grounds that they
were stereochemically very improbable.
They believed that
the following system represents the nearest to truth in the
matter.
N
.sa\
,/
60
61
62
Meuwsen, Berichte 6AB. 2301?i 2311 (1931).
Ibid.. 2303.
Arnold, Hugill, and Hutson, op. cit.. 16A5.
21
The reasons which Arnold, Hugill, and Hutson present
are very concise.
In addition to the hydrolysis in cold di­
lute alkali, which was said to produce ammonia quantitatively,
complete reduction yields no trace of hydrazine.
This fact
was not taken as conclusive.evidence that nitrogen to nitro- gen links do not exist,' but merely as an indication.
They
assumed that the addition of metallic polychlorides was ac­
complished through the unique sulfur atom present in the third
phase above.
Further supporting evidence cited in favOr of
the above system was that in all known sulfur to nitrogen
links, the, polarity has been from the sulfur to the nitrogen,
while in the case of nitrogen to oxygen links, the polarity
has been from the nitrogen to the oxygen.
Thus, a reason is
advanced why sulfur nitride is a tetramer, but nitric oxide
is a monomer.
Arnold et al. do not consider the molecule to
be in a single plane, but rather puckered in much the same
63
fashion as sulfur in Sg. Phalnikar and Bhide
were reported
to have determined the dipole moment of sulfur nitride as
being 0.72 debyes.
They were quoted as interpreting this to
indicate that the third phase proposed by Arnold et al.
is correct.
II.
STRUCTURE SUGGESTED BY THIS WORK
Experimental evidence gathered in this study led to
N. Phalnikar and B. Bhide, Current Sci. 8 , 473 (1939).
22
the conclusion that one of the nitrogen atoms in sulfur nitride
is unique.
Other considerations indicate that no bonding ex­
ists between any two atoms of the same element.
Evidence which
will be discussed in detail in a later section points to an
oxidized form for one of the nitrogen atoms.
The structure
proposed below satisfies the requirements set up by knowledge
gained in this investigation and harmonizes well with informa­
tion obtained by other authors.
It does not necessarily pre­
tend to be correct nor is it intended to add confusion to the
literature, but it seems fully worthy of consideration.
:s:
II
N
/ \
tS: S:
7* I II
':N N:
^ /
S
III
N
CHAPTER V
.EXPERIMENTAL •
I.
PREPARATION OF SULFUR NITRIDE
Following the precise directions of Meuwsen^,
6z.
an at-
tempt was made to prepare sulfur nitride by the standard
method of passing ammonia into a solution of sulfur dichloride
in benzene.
After filtering, the solvent was removed by dis­
tillation, either at room pressure or in vacuo (with or without
a capillary inlet).
The main product always was the red, mal65
odorous oil described by Arnold et al.
and the yield of
sulfur nitride always was low.
It was possible, however, to
obtain enough for a few preliminary experiments.
The best
yield (5 per cent) came from an experiment in which the re­
action went forward in a relatively bright light.
The electrical discharge method of Moldenhauer and
Zimmermann
66
was next tried.
This procedure failed to produce
any observable sulfur nitride even though the discharge was
continued for several days.
An attempt was made to duplicate the work of Ruff and
67
Geisel
using their process of dissolving sulfur in liquid
^ Meuwsen, Berichte 65B. 1724 .
65
■
,
Arnold, Hugill, and Hutson, op. cit.. 1645.
66
^
Moldenhauer and Zimmerman, o£. cit.. 2390 .
Ruff and Geisel, Berichte 38. 2659.
zu
ammonia.
As pointed out previously, the diversity of side
reactions encountered is sufficient to reduce the yield pro­
hibitively.
This trial likewise met with little success.
It was considered possible that a source of nitrogen
other than ammonia might avoid the destructive side reactions
occurring with excess ammonia.
as such a source.
Magnesium nitride was chosen
Although it was treated with sulfur di­
chloride under a variety of conditions, no reaction could be
produced.
In default of success with the more direct methods, it
seemed best to find a way to improve yields from the ammonolysis of sulfur dichloride.
Since sulfur nitride is known to
react with both ammonia and sulfur dichloride, a process was
developed such that no excess of either reactant was present.
At the same time, the use of a liquid solvent and a subsequent
troublesome separation was eliminated.
The procedure followed was to allow liquid ammonia to
evaporate slowly into one side of a large three-necked flask.
The rate of flow was controlled by adjustment of the height
of a cooling bath of solid carbon dioxide and ether which was
placed around the tube containing the liquid ammonia.
Sulfur
dichloride was carried into another part of the flask by means
of a stream of dry nitrogen which served also to dilute the
reaction mixture.
A long tube of large diameter was attached
vertically to the central neck in order to provide a surface
25
i
[
i
j
I
DIAGRAM
OR PREPARATION ASSEMBLY
26
for the precipitation of the large amount of solid material
produced.
The rates of flow of the gases were adjusted so that
a piece of moist litmus paper held at the top of the large
tube indicated a very slight excess of ammonia.
The actual
rate of flow was estimated to be 2 cc per second.
was cooled by a surrounding bath of ice water.
The flask
About twice
the theoretical amount of ammonia was employed in the process.
At the end of the addition time, which was twenty-four hours
for 5A grams of ammonia, the ammonium chloride produced was
removed by the addition of several 500 cc portions of dis­
tilled water.
suction filter.
The solution thus formed was decanted upon a
The resulting deep green solid cake was al­
lowed to stand in air until moderately dry.
II.
PURIFICATION BY VACUUM-SUBLIMATION
The final separation was accomplished by sublimation
of the solid in a tube connected to a high vacuum pump.
tube was heated by a surrounding boiling water bath.
This
While
thus under the combined influence of heat and reduced pressure,
the sulfur nitride sublimed slowly, but cleanly away from the
residue consisting chiefly of sulfur.
This method of puri­
fication, sublimation at the lowest possible pressure and
temperature, was developed by Meuwsen and Kruger.
68
68
A. Meuwsen and S. Kruger. Z. anorg. allgem. Chem.
236. 221 (1938).
~ ---------
27
Burt^ ana Usher?® both found that when sulfur nitride
is sublimed in the presence of sulfur, a substance of the
formula NS 2 was produced.
However, they did not employ the
high vacuum technique, which seems very important.
A small
amount of volatile red oil was in evidence during the vacuum
sublimation, but since it was more volatile than ordinary sul­
fur nitride, it did not contaminate the principal product.
The oily substance may have corresponded to the formula N 2S 5
mentioned by many previous authors.
A resublimation of the
sulfur nitride thus purified, produced bright orange crystals
with an adamantine luster.
A colorless oil accompanied this
product even when the high vacuum method was used.
removed by drying on a filter paper.
It was
The solid detonated
vigorously when struck with a hammer and burned with a blue
flame when ignited.
The yield of sulfur nitride obtained by this method
was 25 per cent.
Although this did not compare favorably with
the yield reported by Van Valkenberg and Bailar?-1 (65 per
cent), the method was adopted, since it seemed to give greater
assurance of a product as well as being a neater method of
handling the substances involved.
69 f # Burt, Froc. Chem. Soc. 26. 127 (1910), Journ.
Chem. Soc.
117l J l 9 1 0 T .
?° Usher, ojd. cit.. 7 3 0 .
Van Valkenberg and Bailar, o£. cit., 2134.
28
III.
ANALYSIS OF THE FRODUCT
The final product hacl a decomposition point of 164 .8°C.
Several analyses gave an average value for the nitrogen of
23.38 per cent.
Inasmuch as the theoretical value is 30 .U U
per cent, this discrepancy is very closely equivalent to one
nitrogen atom.
The method followed was to trap the ammonia,
released by alkaline hydrolysis, in a saturated boric acid
solution, with subsequent titration by a standard acid.
Initial digestion with sulfuric acid is unsatisfactory since
violent dscomposition occurs on warming.
The addition of
granulated aluminum to the alkaline hydrolysis solution pro­
duced a steady evolution of hydrogen and brought about the
desired effect of releasing the fourth atom of nitrogen.
Thus
a far reaching conclusion was made; three of the nitrogen
atoms appear to be ammoniacal in character, while the fourth
is in a higher state of oxidation.
The analysis for sulfur was made in a rather unortho­
dox manner.
Alkaline hydrolysis always produced a small
amount of free sulfur which could not be easily oxidized by
treatment with bromine or hydrogen peroxide.
Consequently,
the policy was adopted of allowing the hydrolysis to go to
completion, oxidizing the dissolved sulfur by hydrogen
peroxide and bromine, filtering and weighing the free sulfur
and then precipitating the oxidized sulfur in the usual
manner.
The combined percentages gave an average value of
29
65.8 per cent as compared with a theoretical value of
69.56 per cent.
IV.
PREPARATION OP THE BORON HALIDES '•
'The boron trichloride used in this work was prepared
by passing a stream of chlorine gas over calcium boride at
800°C.
At the time this was done, the more convenient method
of Gamble, Gilmont and Stiff^ 2 had not yet been published.
Boron trifluoride was prepared by the usual method, from
73
potassium fluoborate, boric oxide, and sulfuric acid.'^
The
boron trichloride was purified in a high vacuum system by
fractional condensation, collecting the portion which was
trapped at -110 °C.
The vapor pressure was measured at 0 °C
to be 487.5 mm, a value which agreed well with that of Stock
and Kuss?^ when corrected for the difference in the values of
the gravitational constant.
in a like manner.
Boron trifluoride was purified
The sample used in this work exerted a
pressure of 293 mm at -112 .5°C.
The boron bromide was pre­
pared by the method of Gamble et al.75 ancL exerted a vapor
72 e # Gamble, P. Gilmont, and J. Stiff, I. Am. Chem.
Soc. 62, 1257 (1940).
73
H.
21-24 (1939).
^ **
75
Booth and K. Wilson, Inorganic Synthesis I
H. Booth, Editor in.Chief, McGraw Hill Book Co.
A. Stock and E,; Kuss, Berichte. 56b, 1463 (1923).
Loc. cit.
30
pressure of 18 mm at 0 °C.?^
As a preliminary experiment, gaseous boron trichloride
was passed into a solution of sulfur nitride in carefully
dried carbon tetrachloride.
No reaction was observed.
A
weighed .sample of sulfur, nitride was next .sealed in a bomb
tube containing an excess of boron trichloride, but the
difficulties of quantitatively determining the extent of a
reaction under such circumstances were numerous.
It was soon
obvious that a more effective method of handling the materials
had to be adopted. -The high vacuum technique suggested itself
as being the simplest and most versatile.
V.
APPARATUS AND TECHNICAL METHODS
The high vacuum technique as here employed is like
that developed by Stock?? for the purpose of working with
the hydrides of boron and silicon.
His original technique
has been modified by-many later workers.
In this study
mercury float valves and a mercury manometer were used until
it was found that certain reaction products reacted vigor­
ously with mercury.
Due to this contingency, it was found
necessary to' substitute an all glass system.
The Jackson
?6 Stock and Kuss, op,, cit.. 1463.
?? A. Stock, Hydrides of Boron and Silicon. Appendix
Cornell University Press, Ithaca, New York (1933).
31
78
mouification of a Bourdon, gauge
was used as a null instru­
ment in combination with a mercury manometer for pressure
measurements.
An all-glass valve controlled by two electro­
magnets was developed from an original idea by C . E. Lane
79
which was later modified by Burg.
A diagram of the appara­
tus as used and the detail of the all-glass valve follow.
The reaction tube A, fitted with a ground glass joint
for easy removal and an electromagnetic stirrer, was at the
extreme left of the system.
The Jackson gauge B was connected
to a XT tube whose volume was accurately known.
The external
jacket of the gauge was connected through a stopcock to a
pump.
A three-way stopcock was arranged to allow either a
rapid or a slow flow of air to enter for balancing purposes.
The groups of U tubes F and G were provided for the purpose
of separating mixtures of gases by fractional condensation.
The small tube H and the large bulb I served for the storage
of boron trichloride and boron trifluoride respectively.
The
tube J was used for weighing the samples of sulfur dioxide
which was used as a solvent. The tube opener K, like that
80
developed by Stock,
was used for the introduction of volatile
samples.
78
T. Phipps, M. Speedmen, and T. Cooke, J. Chem.
Education. 12, 321 (1935).
~
79
A. Burg, Private Communication.
80
Stock, 0£. ext.. Appendix.
iI
I
Mmeupy Flom iAivis
Au.-Gl.ASS, PfAGNCT ic Va lv c
M
I
H i g h Va c u u m Ap p a r a t u s
4.
33
L
e ft
H
alf
Al l -G
lass,
Ma
g n e t ic
Va l v e
To R i g h t H a l f
34
One advantage of the all-glass valve developed during
.this work is that it will remain open or closed without the
continuous use of the electromagnets.
Since these may be
removed without closure of the valve, it is possible to pass
difficultly volatile materials through by flaming.
The valve
is opened by placing an electromagnet around the vertical
tube.
While the valve stem is thus raised, another external
electromagnet is placed about the horizontal tube containing
the valve support 0.
The latter is forced toward the valve
to engage the support collar on the valve stem.
The current
through both magnets now is shut off and the coils are with­
drawn.
This process is repeated on the other half of the
valve.
In order to close the valve, the procedure is reversed.
VI.
ACTION OF ELECTRON ACCEPTORS
Preliminary experiments. Previous to the construction
of the apparatus as diagrammed four preliminary experiments
were performed.
The first of these consisted of weighing a
sample of sulfur nitride into the reaction tube, after which
a measured quantity of boron trichloride was condensed upon
it.
These substances were allowed to stand in contact at
various temperatures for some time, but no reaction was ob­
served.
In the second experiment, purified carbon tetra­
chloride was used as a solvent, but again there were no evi­
dences of reaction between the sulfur nitride and the boron
chloride.
In a third experiment, a sample of boron trifluoride
was placed in contact with a fresh sample of. the nitride.
On
warming to room temperature, the latter suddenly decomposed,
producing a gas which was. not wholly condensable at liquid
air temperatures.
The sample of sulfur nitride taken was
0.001078 moles and the amount of gas produced was 0.00207
moles.
Since nitrogen was the only gas which might properly
have been expected under such circumstances, it was assumed
that the gas was nitrogen.
This assumption was confirmed by
the numerical results, as the complete decomposition of one
mole of sulfur nitride should theoretically produce two moles
of nitrogen.
A considerable proportion (0.00259 moles) of
boron trifluoride was consumed in the process.
81
Martin,
in studying a somewhat similar system, namely
the action of boron trifluoride on hexamethylene tetramine,
had employed sulfur dioxide as a solvent for the* solid.
In
order to subdue the reaction and also to produce a fresh sur­
face for a possible reaction, sulfur dioxide was chosen as a *
solvent here.
Sulfur nitride is relatively quite soluble in
sulfur dioxide even at low temperatures and no reaction be­
tween the two could be detected.
reported
81
Tests, in addition to those
by Martin, were made to insure that no reaction
L. Martin, "The Behavior of Hexamethylene Tetra­
mine Toward Boron Fluoride,” (unpublished Master*s thesis,
The University of Southern California, Los Angeles, 1940).
36
occurred between the solvent and boron trifluoride.
In the fourth experiment of the series, (conducted in
a vacuum apparatus of older design) the use of sulfur dioxide
resulted in the absorption of the initial sample of boron
trifluoride.
Additional samples were also completely absorbed.
This process continued until the molecular ratio of boron
trifluoride to sulfur nitride was 1 to 10.
It then became
obvious that the clouding of the manometer, which at first had
been attributed to impurities in the mercury, was due to some
reaction product, possibly sulfur tetrafluoride.
It was at
this point that the change was made to the all-glass valve
and pressure gauge.
From subsequent measurements, the con­
clusion was reached that a side reaction of small proportions
had been displaced by the presence of mercury, which reacted
with one of the products.
Experiments with the final apparatus. Under the new
conditions, boron trifluoride was absorbed by sulfur nitride
until a maximum ratio of 1.6 to 1 was reached, but was diffi­
cult to reproduce.
of no avail.
The attempts to increase this ratio were
Even complete removal of the solvent and excess
boron trifluoride with the substitution of fresh samples of
these materials did not serve the purpose,
ifust why this
action did not bring about the same effect as resulted from
the presence of mercury is not evident.
The compound thus formed was very readily decomposed
37
by the moisture in air and apparently sulfur nitride is re­
generated.
Direct contact with water resulted in the libera­
tion of a large amount of heat and numerous decomposition
products were formed.
In vacuo, the compound begins to de­
compose to. yield boron fluoride at 45° C, but the process is
not completed except by heating to 100° C.
The residue ap­
pears to differ from the original sulfur nitride in that its
capacity to absorb boron fluoride is considerably diminished.
The exact procedure followed during a typical deter­
mination -is now given.
The apparatus was evacuated to a
pressure of 10 ^ mm, after which a stream of dry nitrogen was
allowed to enter.' The reaction tube was disconnected, stop­
pered, and its adhering picein wax removed by carbon tetra­
chloride.
A sample of pure sulfur nitride was carefully
weighed into the tube which was then reconnected to the system
and evacuated.
A sample of sulfur dioxide was introduced
through the tube opener and purified by distillation from a
bath at -40° C, to a U tube cooled to -80° C, beyond which
was a liquid air trap.
The material caught in the middle bath
exerted a vapor pressure of 120 mm at -45 ° C , corresponding to
82
■
the pure material.
It was transferred to the weighing tube
by condensation at -185° C.
The weighing tube now was removed,
inverted to drop the mercury against the stopcock, and allowed
c>2
A. Stock, Z. Elektrochem. 29, 354 (1921),
38
to warm to room temperature.
A strong rubber band was used
to bold the plug in place during the weighing.
After weigh­
ing, the tube was again attached in its original position;
the system was re-evacuated.
The sulfur dioxide was refrozen
(to allow the mercury to fall away from the stopcock) and
then distilled into the reaction tube.
The evacuated weigh­
ing tube was now reweighed to provide an additional check.
A carefully measured portion of boron trifluoride was then
moved to the reaction tube by means of liquid air.
The whole
was allowed to warm to a temperature slightly above the melt­
ing point of sulfur dioxide (-72° C).
was a liquid at -80° C.
The reaction mixture
It was stirred for some time with
the aid of the magnetic stirrer, after which the sulfur dioxide
and any excess boron trifluoride was slowly distilled from the
o
o
reaction tube at -80 C, toward an adjacent U tube at -120 C
(to trap the sulfur dioxide) and a third U tube at -185° C
(to receive the boron trifluoride).
The purity and quantity
of the sulfur dioxide were rechecked by measuring its vapor
pressure and by reweighing.
The pure boron trifluoride was
carefully measured in the same manner as the original sample
and the amount absorbed was calculated from the two values.
A fresh sample of boron trifluoride was added in the same
manner and any further absorption was noted.
In order to make certain that a foreign gas of a
volatility similar to that of boron trifluoride had not been
produced, advantage was taken of the fact that the latter
39
forms an almost non-volatile compound with diethyl ether.
Very small samples of ether, which had been purified in a man­
ner similar to that used for sulfur dioxide, were carefully
measured and allowed to mix with the recovered boron trifluo♦
V
ride.
The volume was redetermined and, as expected, the
volume decrease was found to equal the volume of ether added.
This process was repeated until a volume of negligible pro­
portions was reached, after which further additions of ether
caused an increase in the volume.
At least part, if not all,
of the material represented by the minimum volume could have
been ether.
At least it was shown that any gaseous products
formed were of negligible proportions.
In other instances, to provide a final check, the
reaction tube was removed, weighed and after the material in
it had been removed by washing with water, the tube was re­
weighed.
In later determinations in which boron trichloride
was used, a gravimetric determination of the per cent chlorine
was made.
Thus, three sources of values were used: (1 ) volu­
metric measurement of the gases involved; (2 ) determination
of the weight increase of the reaction tube; (3 ) analysis of
the solution obtained by hydrolysis of the compound formed.
As will be seen in the tabulated data, in any one instance
only two of the above sources of values were used.
The following tables summarize the experimental data
for both the boron fluoride and boron chloride addition
compounds.
40
TABLE I
BORON FLUORIDE— SULFUR NITRIDE RATIO
Experiment I
Experiment II
Weight S^N^
0.0868 gms.
0.0812 gms.
Moles of S^N^
0.000472
0.000441
Weight of SO 2 added
0.626 gms.
1.3183 gms.
First vol. BF3 added
32.5 cc
20.85 cc
First vol. BF3 recovered
19.0 cc
10.84 cc
Second vol. BF3 added
30.2 cc
Second vol. BF3 recovered
26.8 cc
Total BF3 added
62.7 cc
20.85 cc
Total BF3 recovered
45.8 cc
10.84 cc
Total BF3 absorbed
16.9 cc
10.0
cc
Moles BF3 absorbed
0.000755
0.000446
Ratio BF3 absorbed to S.N^
from volume measurements
1.6 : 1
1.01
Vol. BF3 returned by
heating to 100°C
15*0 cc
10.4
:1
cc
Moles BF3 returned by
heating z o 100°C
0.00067
0.000464
Ratio BF3 to S4N4 from
decomposition study
1.42 : 1
1.05 ; 1
4i
TABLE II
BORON CHLORIDE— SULFUR NITRIDE RATIO
Experiment
I
Experiment Experiment
II
III
Weight S^N^
0.0829
0.3342
0.0812
Moles S/N^
0.000451
0.00182
0.000441
Weight SO2 added
3.2540
1.4119
1.3024
Weight S02 recovered
3.254
1.4083
1.3022
First vol. BCI3 added
37.9 cc
15.5 cc
22.9 cc
First vol. BCI3 returned
21.4 cc
Second vol. BCl^ added
10.6
36.8 cc
Second vol. BCl^ returned
0.00
Total BCl^ added
37.9 cc
Total BCl^ recovered
21.4
0.0
10.6
Total BCl^ absorbed
16.5
52.3
12.3
52.3 cc
22.9 cc
Moles BCl^ absorbed
0.000737
0.00233
0.000549
Ratio BCl^ to S^N^ from
volume measurements
1.6 : 1
1.3:1
1.3 : 1
Weight addition compound
0.5562
Ratio BClo to S 1N7 from
weighing
1.04 : 1
Weight AgCl pptd. from solu­
tion of addition compound
0.2703
0.2183
Ratio BC1-* to
analysis
1.4 : 1
1.2 : 1
from
42
Before an attempt to use boron bromide as an additor,
its behavior toward sulfur dioxide was tested.
rapid at room temperatures, was found to occur.
A reaction,
The princi­
pal products seemed to be sulfur bromide, bromine and a white
solid which was assumed to be boric anhydride.
As a result
of this knowledge, carbon tetrachloride was substituted for
sulfur dioxide as a solvent during the work with sulfur
nitride and boron bromide.
The action of boron bromide upon sulfur nitride in the
absence of a solvent resulted in the formation of a dark brown
mass of uncertain composition, accompanied by sulfur bromide
>
and bromine.
In the presence of carbon tetrachloride as a
solvent, boron bromide did not react with sulfur nitride under
ordinary, conditions.
VII. STABILITY OF THE ADDITION COMPOUNDS
As a variation of the above methods of analysis and
as a means of testing the stability of the mixtures formed,
the tube containing the compound was surrounded by a bath,
the temperature of which was gradually raised until decomposi­
tion occurred.
Initial-decomposition of both the boron tri­
fluoride and boron trichloride addition products occurred at
45° C.
Heating at this temperature for ten minutes caused
the production of a quantity of gas which, it was estimated,
decreased the boron halide to sulfur nitride ratio to very
43
close to 1:1.
Although this did not appear to be a halt
point, the removal of boron fluoride became more difficult.
In order to remove all of the absorbed gas, it was necessary
o
to decrease the bath temperature to 100 C. This process
allowed the original quantities of gas absorbed to be re­
checked in the absence of sulfur dioxide.
Attempts to reform
the addition products after such a decomposition were not
completely successful as only one half of the boron halide
could be recombined, even in the presence of sulfur dioxide.
To obtain further information regarding the relative
stabilities of the two products prepared, each mixture was
treated with the other reagent, both with and without the
^ +
solvent, sulfur dioxide. That is, the material S4N 4..I BCL3
was prepared and treated with boron trifluoride and similarly
the reverse was done.
further addition.
In neither case was any evidence of
CHAPTER VI
DISCUSSION
I.
DISCUSSION OF EXPERIMENTAL RESULTS
The action of boron trifluoride and boron trichloride
+
upon sulfur nitride to produce materials of the type S^N^.l BX3
is interpreted as follows: One mole of sulfur nitride reacts
with one mole of the boron halide to form a definite addition
compound; the extra fraction of a mole of the boron halide
represents a solid solution effect in which a very weak sec­
ondary bonding occurs.
The original addition also is rather
weak, and the existence of the 1:1 compound therefore is dif­
ficult to prove beyond doubt.
The assumption of the 1:1 com­
pound implies that there is a unique nitrogen atom in sulfur
nitride, which is somewhat ammoniacal in character.
The. release, of only three of the four nitrogen atoms
by hydrolysis is entirely contradictory to the evidence of
all previous investigators, of the simple hydrolysis.
In the
present study of the alkaline hydrolysis, (without reducing
agent) every effort was made to find the fourth nitrogen atom
as ammonia, but only three could be found.
If this piece of
evidence is accepted, one is forced to the conclusion, either
that there is another unique nitrogen atom in sulfur nitride,
or that upon hydrolysis a mutual oxidation and reduction oc­
curs.
None of the previously suggested formulae contain even
one nitrogen atom which occupies a unique position, yet this
experimental evidence seems to require two such atoms.
The
second atom would he expected to be in a rather highly oxi­
dized state.
The following structure not only satisfies these
requirements, but agrees well with the studies of previous
investigators.
N
lit
S.
II
s
The indicated internal dative bonding would serve to
explain the limitation of the formula to
and the absence
of other polymers, for these internal bonds would suppress
tendencies toward linkages to other molecules.
The weakness
of the boron halide linkage to the uppermost nitrogen atom
thus is easily understood.
As neither sulfur dioxide nor
hydrogen sulfide react with boron trifluoride, and the oxygen
atom of the radical N=0, although a very poor boron trifluoride
receptor yet is still better than sulfur in a similar state,
the attachment of the boron halides to sulfur nitride is al­
most certainly through a nitrogen atom.
The structure sug­
gested would in all probability undergo resonance in much the
same fashion as benzene.
46
It is not suggested that the above formula is the
only possible structure which would have two unique nitrogen
atoms in suitable states of oxidation, for indeed others are
possible. -However , -strained bond ■angles -must bje avoided and
the new structure here suggested ^at least seems ^worthy- of
further consideration..
BIBLIOGRAPHY
A.
BOOKS
Booth, H. S., and K. S. Wilson, Boron Trifluoride. H. Booth,
editor, Inorganic Synthesis I; New York: McGraw-Hill Book
Company, 1939.
Stock, A., Hydrides of Boron and Silicon,
Cornell University Press, 1933.
B.
Ithaca, New York:
PERIODICAL ARTICLES
Arnold, M . , Hugill, J., and Hutson, J., J. Chem. Soc. 1645
(1936).
~
Bergstrom, F., Journ. Am. Chem. Soc. 4 S
.
2319 (1926).
Buerger, M . , Am. Mineral. 21, 575 (1936).
Burt, F,, Proc. Chem. Soc. 26. 127 (1910).
_______, J. Chem. Soc. 2Z, 1171 (1910).
Clever, A., and Muthmann, W., Berichte 29. 340 (1896).
De Marcay, E., Comptes Rendus 91. 854, 1066 (1880).
Davis, 0. C., J. Chem. Soc. 8£, 1575 (1906).
Fordos, J., and Gelis, A., Comptes Rendus 31. 702 (1850).
Francis, F., and Davis, 0., J. Chem. Soc. 85. 259 (1904).
Gamble, E., Gilmont, P., and Stiff, J., J ourn. Am. Chem. Soc.
62, 1257 (1940).
Gregory, Jo urn. Pham. 21, 315, 22, 301 (1835).
Jaeger, F., and Zanstra, J., Proc. Acad. Sci. Amsterdam, 34.
782 (1931).
Lengfeld, F., and Stieglitz, J., Berichte 28. 2742 (1895).
Meuwsen, A., Berichte 62 B. 1959 (1929).
_______, Berichte 64 B. 2301, 2311 (1931).
Meuwsen, A*, Berichte 65 B, 1724 (1932).
Meuwsen, A., and Kruger, S., Z. anorg. u. allgemein. Chem.
236. 221 (1936).
Moldenhauer, W. , and Zimmerman, A., Berichte 62 B. 2390 (1929).
Phalnikar, N., and Bhide, B., Current Sci. 8., 473 (1939).
Phipps, T., Spealman, M., and Cooke, T., J. Chem. Education
12, 321 (1935).
Ruff, 0., Z. anorg. u. allgemein. Chem. 207. 242 (1932).
Ruff, 0., and Geisel, E., Berichte 37. 1573 (1904).
Berichte 38. 2659 (1905).
Schenck, R., Naturforschenden Gesellschaft zu Halle a.S.
February (18955.
______ , Ann, der Chimie 290. 171 (1896).
Soubeiran, Ann. Chim. Phys. 67 . 71, 96 (1836).
Stock, A., and Kuss, E. .""Berichte 56 B. 1463 (1932).
IJsher, F., J. Chem. Soc. 127 . 7 30 (1925).
Van Valkenberg. H., and Bailar, J., Journ. Am. Chem. Soc. 4 7 .
2134 (1925).
Voznesenskii, S ., Journ. Russ. Phys. Chem. Soc. 59. 221
(1927).
WSlbling, H., Z. anorg. Chem. 52., 281 (1908).
C . PATENTS
Swinehart, C. F., U. S. Patent 2,190,177, February 13 (1940).
D.
UNPUBLISHED MATERIAL
Martin, L. I., "The Behavior of Hexamethylene Tetramine Toward
Boron Fluoride." Unpublished Masterfs thesis, The
University of Southern California, Los Angeles, 1940.
Документ
Категория
Без категории
Просмотров
0
Размер файла
2 288 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа