Fluorine has got -1 as always, rendering Oxygen and Hydrogen +1/2 each. Oxygen shows non integral Oxidation Valency state in some compounds such as KO2 and Na2S4O6. But I am not aware of Hydrogen having a non integral valency. Any thoughts?
'Zero' is my feeling (whilst recognising oxidation states are just formalisms):
The O-F bond is polarised O+-F-
The H-O bond is polarised H+-O-
The + and - on oxygen cancel out leaving zero
Yes that is a possibility too because having Hydrogen with +1/2 is very unlikely ! I have never heard of anything!
OH is isolectronic with F, hence HOF is comparable with F2, wheRe F is zerovalent as well as the O and H atoms in HOF. Obviously the different electronegativities may determine some polarization of the linkages but not enough to change the formal oxidation states.
According to your discussion, please find below quantum chemical NBO analysis of HOF and system (F-)x(H2O) (at ambient conditions).
Geometry parameters:
H-O (0.942020 Angstr.), O-F (1.360199 Angstr.), HOF (angle) 108.900352 deg.
Population:
If you have only F-, that the e-configuration should be 1s22s2 2p6 or 10e. For (F-)xH2O system, it has been obtained "9.97883". So that the results have supported the common theory.
But forHOF the value is "9.13832".
Therefore you have
H(delta1+)-O(delta2-)-F(delta3-) system, The data rather support the commentsof Mr. Mealli, than those ones of Mr. Cade.
SYSTEM: HOF
%Chk=HOF.chk
------------------------------------------
# UB3LYP/6-311++G(d,p) Freq Pop=(NPA) Test
------------------------------------------
1/10=4,30=1,38=1/1,3;
2/17=6,18=5/2;
3/5=4,6=6,7=1111,11=2,25=1,30=1/1,2,3;
4/7=2/1;
5/5=2,38=4,42=-5/2;
8/6=4,11=11,23=2/1;
11/6=1,8=1,9=11,15=111,16=11/1,2,10;
10/6=1/2;
6/7=2,8=2,9=2,10=2,18=1,28=1,40=-1/1,7;
7/8=1,10=1,25=1/1,2,3,16;
1/10=4,30=1/3;
99//99;
---
HOF
---
Symbolic Z-matrix:
Charge = 0 Multiplicity = 1
O 0 0.33238 0.36786 -0.09866
F 0 -0.90888 -0.09082 0.21603
H 0 0.90888 -0.36786 -0.21603
GradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGrad
Berny optimization.
Initialization pass.
Trust Radius=3.00D-01 FncErr=1.00D-07 GrdErr=1.00D-07
Number of steps in this run= 20 maximum allowed number of steps= 100.
GradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGrad
------------------------------------------------------------------------
Z-MATRIX (ANGSTROMS AND DEGREES)
CD Cent Atom N1 Length/X N2 Alpha/Y N3 Beta/Z J
------------------------------------------------------------------------
1 1 O 0 0.332382 0.367859 -0.098663
2 2 F 0 -0.908875 -0.090820 0.216034
3 3 H 0 0.908875 -0.367859 -0.216034
------------------------------------------------------------------------
................................................
Natural Population
Natural -----------------------------------------------
Atom No Charge Core Valence Rydberg Total
-----------------------------------------------------------------------
O 1 -0.31188 1.99977 6.28927 0.02285 8.31188
F 2 -0.13832 1.99997 7.12368 0.01467 9.13832
H 3 0.45020 0.00000 0.54516 0.00464 0.54980
=======================================================================
* Total * 0.00000 3.99974 13.95811 0.04216 18.00000
................................................
Atom No Natural Electron Configuration
----------------------------------------------------------------------------
O 1 [core]2s( 1.77)2p( 4.52)3s( 0.01)3p( 0.01)3d( 0.01)
F 2 [core]2s( 1.91)2p( 5.22)3s( 0.01)3p( 0.01)
H 3 1s( 0.55)
................................................................................
Summary of Natural Population Analysis:
Natural Population
Natural -----------------------------------------------
Atom No Charge Core Valence Rydberg Total
-----------------------------------------------------------------------
O 1 -0.15594 0.99989 3.14463 0.01142 4.15594
F 2 -0.06916 0.99998 3.56184 0.00734 4.56916
H 3 0.22510 0.00000 0.27258 0.00232 0.27490
=======================================================================
* Total * 0.00000 1.99987 6.97905 0.02108 9.00000
Natural Population
--------------------------------------------------------
Core 1.99987 ( 99.9935% of 2)
Valence 6.97905 ( 99.7008% of 7)
Natural Minimal Basis 8.97892 ( 99.7658% of 9)
Natural Rydberg Basis 0.02108 ( 0.2342% of 9)
--------------------------------------------------------
Atom No Natural Electron Configuration
----------------------------------------------------------------------------
O 1 [core]2s( 0.88)2p( 2.26)
F 2 [core]2s( 0.95)2p( 2.61)
H 3 1s( 0.27)
BO analysis skipped by request.
Full mass-weighted force constant matrix:
Low frequencies --- -797.4175 -276.4760 -211.6118 -0.0015 -0.0008 0.0006
Low frequencies --- 1163.3115 1289.0642 4123.0887
Harmonic frequencies (cm**-1), IR intensities (KM/Mole),
Raman scattering activities (A**4/AMU), Raman depolarization ratios,
reduced masses (AMU), force constants (mDyne/A) and normal coordinates:
1 2 3
A' A' A'
Frequencies -- 1163.0129 1287.1261 4123.0538
Red. masses -- 16.2210 1.1092 1.0709
Frc consts -- 12.9270 1.0827 10.7265
IR Inten -- 12.0954 36.7003 45.2647
Raman Activ -- 0.0000 0.0000 0.0000
Depolar -- 0.0000 0.0000 0.0000
Atom AN X Y Z X Y Z X Y Z
1 8 0.01 0.74 0.00 -0.06 -0.03 0.00 0.06 -0.02 0.00
2 9 0.00 -0.63 0.00 0.04 -0.03 0.00 0.00 0.00 0.00
3 1 -0.23 0.11 0.00 0.24 0.97 0.00 -0.95 0.29 0.00
SYSTEM: (F-)xH2O
%Chk=Untitled-5.chk
------------------------------------------
# UB3LYP/6-311++G(d,p) Freq Pop=(NPA) Test
------------------------------------------
1/10=4,30=1,38=1/1,3;
2/17=6,18=5/2;
3/5=4,6=6,7=1111,11=2,25=1,30=1/1,2,3;
4/7=2/1;
5/5=2,38=4,42=-5/2;
8/6=4,11=11,23=2/1;
11/6=1,8=1,9=11,15=111,16=11/1,2,10;
10/6=1/2;
6/7=2,8=2,9=2,10=2,18=1,28=1,40=-1/1,7;
7/8=1,10=1,25=1/1,2,3,16;
1/10=4,30=1/3;
99//99;
----------
Untitled-5
----------
Symbolic Z-matrix:
Charge = -1 Multiplicity = 1
F 0 -1.28641 0.14641 0.1138
O 0 1.28642 -0.04414 -0.05965
H 0 0.6212 -0.70551 -0.11382
H 0 0.73666 0.70549 0.07632
GradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGrad
Berny optimization.
Initialization pass.
Trust Radius=3.00D-01 FncErr=1.00D-07 GrdErr=1.00D-07
Number of steps in this run= 22 maximum allowed number of steps= 100.
GradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGrad
------------------------------------------------------------------------
Z-MATRIX (ANGSTROMS AND DEGREES)
CD Cent Atom N1 Length/X N2 Alpha/Y N3 Beta/Z J
------------------------------------------------------------------------
1 1 F 0 -1.286407 0.146408 0.113800
2 2 O 0 1.286423 -0.044144 -0.059647
3 3 H 0 0.621200 -0.705505 -0.113815
4 4 H 0 0.736664 0.705490 0.076324
------------------------------------------------------------------------
Summary of Natural Population Analysis:
Natural Population
Natural -----------------------------------------------
Atom No Charge Core Valence Rydberg Total
-----------------------------------------------------------------------
F 1 -0.97883 2.00000 7.97829 0.00055 9.97883
O 2 -0.99336 1.99981 6.97871 0.01484 8.99336
H 3 0.48600 0.00000 0.50893 0.00507 0.51400
H 4 0.48619 0.00000 0.50874 0.00507 0.51381
=======================================================================
* Total * -1.00000 3.99981 15.97466 0.02553 20.00000
Natural Population
--------------------------------------------------------
Core 3.99981 ( 99.9952% of 4)
Valence 15.97466 ( 99.8416% of 16)
Natural Minimal Basis 19.97447 ( 99.8723% of 20)
Natural Rydberg Basis 0.02553 ( 0.1277% of 20)
--------------------------------------------------------
Atom No Natural Electron Configuration
----------------------------------------------------------------------------
F 1 [core]2s( 2.00)2p( 5.98)
O 2 [core]2s( 1.74)2p( 5.24)3p( 0.01)3d( 0.01)
H 3 1s( 0.51)
H 4 1s( 0.51)
.............................................................
Summary of Natural Population Analysis:
Natural Population
Natural -----------------------------------------------
Atom No Charge Core Valence Rydberg Total
-----------------------------------------------------------------------
F 1 -0.48942 1.00000 3.98914 0.00027 4.98942
O 2 -0.49668 0.99990 3.48935 0.00742 4.49668
H 3 0.24300 0.00000 0.25446 0.00253 0.25700
H 4 0.24310 0.00000 0.25437 0.00254 0.25690
=======================================================================
* Total * -0.50000 1.99990 7.98733 0.01277 10.00000
Natural Population
--------------------------------------------------------
Core 1.99990 ( 99.9952% of 2)
Valence 7.98733 ( 99.8416% of 8)
Natural Minimal Basis 9.98723 ( 99.8723% of 10)
Natural Rydberg Basis 0.01277 ( 0.1277% of 10)
--------------------------------------------------------
Atom No Natural Electron Configuration
----------------------------------------------------------------------------
F 1 [core]2s( 1.00)2p( 2.99)
O 2 [core]2s( 0.87)2p( 2.62)
H 3 1s( 0.25)
H 4 1s( 0.25)
An interesting analysis Dr Ivanova.
From the calculations it does seem that the oxygen atom carries most of the negative charge (its notable that while the O2s orbital is deficient in electron density the opposing contributions of the 2p orbitals seem to outweigh this effect
O 1 [core]2s( 1.77)2p( 4.52)3s( 0.01)3p( 0.01)3d( 0.01)
F 2 [core]2s( 1.91)2p( 5.22)3s( 0.01)3p( 0.01)
...) I suppose this might be interpreted as the O-F sigma bond being polarised O+-F- but the pi-backbonding from F to O outweighing this.
However, some of the reactivity exhibited by HOF does suggest that the oxygen atom is the electrophillic centre: Chem. Commun., 2013,49, 7379 (DOI: 10.1039/c3cc42337a)
RB(OH)2 + HOF -> ROH (and F-B(OH)2 presumably)
and 19F NMR studies (J. Chem. Phys. 57, 4542 (1972) (doi.org/10.1063/1.1678113)) suggest that the fluorine nucleus in HOF is only a little less electron rich than for the fluorine in HF. (there is also a mention of a contracted H-O-F angle, 97.2 degrees, consistent with an H...F interaction in H+-O-F-)
That being said, earlier IR studies (Spectrochim. Acta 28A, 65 (1972)) suggest a near zero charge on fluorine based on the differences in IR spectra of HOF, F2 and OF(radical).
So, in summary, the actual charges on O and F are probably not that large (consistent with the electronegativities of the two elements not being very different).
But, since oxidation states are entirely a formalism and derived from considering the electronegativities of the atoms at each end of an ionic bond, I would still say H+...O0...F-, though the experimental reality maybe somewhat greyer!
In my opinion, the O-F bond is essentially covalent with the O and F atoms equally sharing one sigma bonding electron each. The O atom is electron richer than fluorine because its orthogonal p orbital involved in the O-H sigma bonding hosts in part H electrons. The idea of a F-O pi-backbonding can be dismissed because the O-H sigma* level lies very high in energy given the strenght of the quasi othogonal O-H linkage, Indeed sigma* is the unique empty level in the system, but it canbe hardly considered an accessible acceptor.
Concerning the HOF reactivity, recall that the hypoflorous acid is explosiveto give O2 + HF. The formation of the O-O linkage seems inconsistent with an initial nucleophilic power of the fluorine atom, as perhaps suggested by the FB(OH)2 formation. Certainly the occurred O oxidation in O2 associates to F reduction, but possibly not as the first step.
May be, it would be an interesting study to determine the reaction profile of 2HOF --> O2 + 2HF. or this has been already done.
I agree, the OF bond is probably not very polarised at all in reality. Indeed, the observed electrophillic character of the 'OH' is still consistent with an overall neural or even anionic oxygen atom.
It is entirely possible that the oxygen atom is electron rich, but possesses an electrophilic region opposite the O-F bond with which the nucleophilic carbon in an alkyl/aryl boronate can react.
Well based on the above calculations by Dr Ivanova and analyses by Ian and Carlo, I feel that teaching Bonding based on Electronegativity must be altogether abandoned. Instead, from school level, VB theory and MO theory must be taught. Coulson's "The shape and structure of molecules" is a good starting point in this regard.
According your further discussion, please find additional information (incl. new computational data), which again support the statements of Mr. Mealli, than those of Mr. Cade, however.
In general, there have numerous both theoretical and experimental studies of HOF and it reactivity in different media and systems H2O, H2O2, CH3CN and more. Particularly for system 2HOF = O2+ 2HF, I have not founds data, but the presented literature research can not be defined as "comprehensive" so that may be there has additional information.
Towards the experimental data, which you have discussed:
1. In the literature have studies on 19F-NMR of HOF and HF in CH3CN, measured at NMR instruments operating with 300 and 250 MHz. But the data have shown
19F = -8.5 ppm (HOF/CH3CN) [ref. 12] and -182.02 ppm (HF/CH3CN) [ref. 9] (new attachment)
According the presented herein calculation of NMR data, however, the
GIAO magnetic shielding tensor components (ppm): iso: 314.91, aniso: 249.8267 (HF); iso: 244.0581, Aniso.: 360.5281 (HOF) have been obtained.
So that to obtain a difference of ca. 183 ppm in 19F and about 2 ppm in 1H-NMRs, assumes that most probably in [ref. 12]
has occurred a product of interaction of HOF/CH3CN, due to the chemical reactivity of HOF such as this one discussed in for example refs. [8 - 13]
2. You have provided structural parameters, which I have found in ref. [2] for example (rOH, 0.966 A; rOF, 1.442 A, and angle FOH 96.8deg.). But the crystallographic data of the HOF (ref. [7]) have shown values (rOH, 0.78 A; rOF, 1.442 A, and angle FOH 101deg.). In this respect we can compare quantum chemical data in for example ref. [2, 7] or herein presented ones only using the r(OF). It is know that there have shifting between theoretical and the experimental data between the different theoretical levels, but the scaling parameters are (or can be) determined, but this shifting is defined.
3. There have also available IR, and Raman data, where can be found some dissonance between the assignment of HOF/DOF [refs. 1 and 5], but all the assignments has shown 889 cm-1 as niOF. This frequency in fact is a constant, which is effect insignificantly from the deuteration, but it is sensible to isotopic exchange H18OF, showing shifting of ca. 40 cm-1 [ref. 5], which is expectable describing the isotopic effect in the IR spectra. Nevertheless that there has abnormally strong intensity in both IR- and Raman spectra. In addition to 1397 cm-1 has been assigned IR - band of d(HOF), which as a result of deuteration should shift to the highfrequencies. In the shown IR-spectrum [ref. 5] for the d(DOF) band has been assigned a frequency ca. 1000 cm-1. Unfortunately there have not literature data about the polarized IR - and Raman analysis, which unambiguously to confirm the shown assignment.
After these comments (points 1-3) on the available experimental set of data, please find quantum chemical NBO analysis of systems [O(0)-F]- and [(F0)-O]-. As you can see, in spite of the initial state, the optimization of the molecular geometry at one and the same level, has resulted to equal data about the charges and the population (Please note that the charges are also shown in the table "Population").
The data have shown again (Odelta4(-)-Fdelta5(-)) (page 3 in the attachment).
4. Now if we focus our attention on the available data for FO-, ClO-, BrO-, IO-, but from photoelectron spectroscopy [ref. 3], it is shown that within the frame of the series the spin-orbit splitting for OF- (193.8 cm-1) is less with factor 5 than the vibrational frequency. For OI- the value is 2091 cm-1.
In this respect, the study [ref. 3] has shown that you have in fact level of covalency of O-F. Furthermore, if we used the shown frequency 769 cm- 1 for FO- as xperimental value for scaling of the theoretical data, a factor 1.5786 is obtained (attachment).
If we use, this factor to compute the shifting of the theoretical spectra of HOF (here previous data), using the theoretical band 1166 cm-1 ban for n(OF), we have obtained a value of 734 cm-1 close to this one of the isolated FO- anion of the experimental photoelectron spectroscopy. Furthermore this is accordance with the theoretical NBO analysis of isolated FO- and HOF (new attachment and previous message).
This result again confirm that HOF can be described as:
H(delta3)(+)-O(delta4(-))-F(delta5(-)) (attachment page 3), which supports as mentioned above the statements of Mr. Mealli!.
Thanks Dr Ivanova, a well presented description of HOF (will have to spend a little more time having a look at all the references!)
However, I think the discussion has slipped a little from the initial question, which was, what are the oxidation states of the individual atoms in HOF? To determine these formal states, the only things you need to consider are the electronegativities of the individual atoms, and assume all bonding is wholly ionic (I am not for one moment suggesting that in reality the O-F bond is even mostly ionic!). So, we need:
(Pauling) Electronegativity of F = 3.98
(Pauling) Electronegativity of O = 3.44
(Pauling) Electronegativity of H = 2.2
And we get:
F-O is considered F- O+
H-O is considered O- H+
Which gives:
F(-1)-O(0)-H(+1)
Oxidation states do not purport to say very much about the chemical reality of any system (as can be seen by the fairly extreme assumptions that are used in their definition). What they do allow you to do is keep track of electrons and define reactions as redox or not.
For another extreme example of oxidation state numbers, methane is considered as follows:
(Pauling) Electronegativity of C = 2.55 (H is still 2.2)
So for each of the C-H bonds the situation is C- H+ which gives the carbon an overall oxidation state of -4 ! (each of the hydrogens is in a +1 oxidation state). However despite this formalism, I don't think there is any doubt that the bonding in methane is somewhat more covalent rather than ionic.
In summary I agree with Jestin. When designing a course on chemical bonding it is best to approach the subject using VB or MO theory rather than oxidation states.
I think that the proposal that in methane C is -4 and H +1 is quiteunrealistic. Also, the transformation of any largely covalent C-H bond into an ionic picture (e.g., C-/H+) is uncorrect and deceiving, especially because a small electronegativity difference (2.55 and 2.22 for C and H, respectively) is inconsistent with the separation of the charges..
Also consider that the electronegativity is usually represented by a unique number for a given atom, which form various linkages with its atomic orbitals of different for symmetry and energy. This implies, that the various bonds have different components depending on their energy and weight in the bonding levels. As an example, CH4 consists of four bonds which are equivalent by symmetry, but one cannot say that also the involved eight electrons have all one energy. The electron pair in the bonding level of a1 symmetry, formed by the a1 Cs orbital, lies deeper than the three degenerate t2 bonding levels formed by the equivalent C px, py and pz orbitals. Hence the character of each bond sums up from different components, each one corresponding to a somewhat different redistibution of the bonding electrons shared by the atoms.
Only by performing a suitable analysis of the wavefunctios, bonding feature are better understood, espeically by using as a reference the different energies of the atomic orbitals rather than the unique electronegativity value. Also, from the latter analysis, some good qualitative understanding of the oxidation states can be extracted, which is a measurable magnitude, but is exploited by the common chemical intuition to guess structural and reactivity features.
In conclusion, I agree with you all that chemical bonding cannot be reduced to one single descriptor.
I apologize for the typo. Obviously the oxidation state of an atom is "not" a measurable magnitude.
- Badges
- Science method
- Similar topics
- Chemistry
- General Biochemistry
- Biochemical Techniques
- Analytical Chemistry