A Matrix Isolation ESR and Theoretical Study of ZnN - ACS Publications

Page 1 of 47. ACS Paragon Plus Environment. The Journal of Physical Chemistry. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. ...
0 downloads 0 Views 604KB Size
Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC

A: Spectroscopy, Molecular Structure, and Quantum Chemistry

A Matrix Isolation ESR and Theoretical Study of ZnN Thomas Sandow Hearne, Sally Ann Yates, Duncan Andrew Wild, and Allan James McKinley J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.9b00601 • Publication Date (Web): 01 Apr 2019 Downloaded from http://pubs.acs.org on April 1, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

A Matrix Isolation ESR and Theoretical Study of ZnN Thomas S. Hearne, Sally A. Yates, Duncan A. Wild, Allan J. McKinley* Chemistry, School of Molecular Sciences, The University of Western Australia, 35 Stirling Highway, Crawley, Western Australia 6009, Australia

*

Author to whom correspondence should be addressed. Electronic

mail: [email protected], Telephone: +61 8 6488 3165.

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT

The Zn14N,

67Zn14N,

Zn15N, and

67Zn15N

radicals have been formed by

the reaction of a plume of zinc metal produced with laser ablation and either ammonia vapor or nitrogen atoms isolated in an inert neon matrix at 4.3 K. The ground electronic state of ZnN was determined to be 4- using electron spin resonance (ESR) spectroscopy. The following magnetic parameters were determined experimentally for ZnN: g⊥ = 1.9998(3), g∥ = 2.0018(3), |D|= 7268(8) MHz, A⊥(14N) = -17.9(20) MHz, A∥ (14N) = 1.5(20) MHz, A⊥(15N) =

25.1(20) MHz, A∥ (15N)=

-2.0(20) MHz, A⊥ (67Zn) = 156(3)

MHz, A∥(67Zn) = 168(12) MHz. The low-lying electronic states of ZnN were also investigated using the complete active space selfconsistent field (CASSCF) technique. By plotting the potential energy surface (PES), theoretical parameters for the ground state with a configuration of 8σ29σ210σ14π2 were determined, including: re = 2.079 Å and De = 1.0 kcal/mol.

ACS Paragon Plus Environment

2

Page 2 of 47

Page 3 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

INTRODUCTION Zinc is an essential element for the growth and health of biological humankind.1

organisms

ranging

from

plants

and

animals

to

There are more than 300 enzymes containing zinc

known.2 Zinc and zinc compounds are also very important reagents in organic synthesis.3

Zinc compounds and alloys are important

members of the group II-VI-type semiconductors and zinc is used as a dopant in group III-V semiconductors.4-5 While in many of these compounds the zinc is present as the divalent cation, with or

without

coordinating

ligands,

covalently-bound

compounds

containing zinc, particularly organometallic derivatives, are also known.

One approach to gain insight into the electronic

structure of such species is to measure the hyperfine coupling in radicals formed from them.

However, very few zinc-containing

radicals have been analyzed in this way.

In this report we

present the results of a matrix isolation ESR study of several isotopologues of the ZnN radical and these experimental results are compared with ab initio calculations involving a range of different levels of theory, and the Potential Energy Surface (PES) for ZnN has been constructed using a complete The ZnN was formed by

active space self-consistent field calculation (CASSCF).

the reaction of laser ablated zinc metal and nitrogen gas, nitrogen atoms and ammonia.

The ground electronic state of ZnN

was determined to be 4- and hyperfine coupling to the Zn-67 and N-14 and N-15 nuclei have been measured. experimental

observation

of

the

ZnN

radical.

ACS Paragon Plus Environment

3

This is the first Similar

4Σ-

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 47

radicals that have been studied using matrix isolation ESR, include MgN6, Sc2+ 7, C2+ 8, Si2+ and Ge2+ 9, BC10, AlC11, GaP+ GaAs+

13,

12,

SiB and SiAl14 , ZnP and CdP15.

Boldeyrev and Simons16 have conducted ab initio theoretical calculations on ZnN.

They found the ground state to be a

4Σ-

radical bound only by a weak van der Waals interaction, with the excited



i

state with a stronger bond. In the

4Σ-

state at the

MP2 level of theory, they found two minima, one at 2.028 Å and another at 4.263 Å.

The short bond minimum was found to be at

a higher energy than the long bond minimum.

At the B3LYP level

of theory, only the short minimum was found, but at the QCISD level of theory, only the long minimum was found. Boldeyrev and Simons recommended further studies to determine the geometry of the ground state of ZnN.

Other zinc-containing radicals which

have been investigated using the matrix isolation ESR technique include ZnH17-18

ZnCH318,

ZnF19, ZnP15 and AgZn20.

For ZnH, ZnP

and ZnCH3 the Zn-67 hyperfine coupling was observed.

The small

zinc molecules, ZnOH21-22, ZnCN23, ZnCCH24, and HZnCH325 have all been studied by rotational spectroscopy in the gas phase while ZnO26, OZnO26, ZnH227, ZnOH21, HZnOH and Zn(OH)228, ZnCH229, HZnCH29, HZnCH330-31, and

ZnNO232

have

also

been

studied

using

matrix

isolation IR spectroscopy and Zn233, ZnHg, MgZn and CdZn34, ZnD2 and HZnCH335 have been studied in matrices with UV and VUV electronic spectroscopy. Additionally, the AlZn molecule was studied by resonant two-photon ionization (R2PI) spectroscopy36

ACS Paragon Plus Environment

4

Page 5 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

and

the

ZnCH3

and

ZnC2H5,37

monomethylcyclopentadienyl,39 molecules

were

studied

fluorescence (LIF).

in

and the

Zn-pyrrolyl,38 Zn-

gas-phase

Zn-

cyclopentadienyl40 via

laser

induced

The universal abundance of zinc is of a

similar level to fluorine and fluorine-containing molecules AlF41, HF42, and CF+43 have all been observed in circumstellar environments.

While a number of small zinc-containing molecules

have been studied in the laboratory no molecule containing zinc has been detected in circumstellar or interstellar space. It is probable

that

zinc

containing

molecules

are

present

in

circumstellar space and are likely to be in a similar form to other previously detected molecules containing heavy atoms, namely XH, XO, XOH, XNC, XCN, XC, XCH, XCCH or XN, where X is an atom heavier than nitrogen. Zinc-containing analogues of these

molecules

have

been

observed

experimentally

above.

ACS Paragon Plus Environment

5

as

shown

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 47

EXPERIMENTAL The design of the matrix isolation equipment utilized in this work has been described in detail in a previous publication44so only a brief description will be given here.

Isotopologues of

the ZnN radical were created by the reaction of a plume of laser ablated metal with a reactant gas flowing through the ablation chamber.

The metal target used was natural abundance zinc (BDH,

LR grade, 99.9% pure). A Nd:YAG laser (Surelite I) was frequency doubled to 532 nm and pulsed at 10 Hz with an energy of around 15 mJ per pulse.

The output from this laser was focused onto

the ablation target to generate a plume of zinc vapor.

Several

reactant gases were used to react with the metal plume to generate the ZnN radical.

A stream of N atoms was formed by

irradiating a 1:2000 mixture of nitrogen gas with neon gas (99.999% Spectra Gases) with a 2,450 MHz microwave power source (modified Microtizer MT-150 Minato Medical Science Co. Ltd) combined with an Evenson cavity (Opthos Instruments) and was successfully used to produce the ZnN radical. Both N-14 natural abundance N2 gas (Industrial Grade, Air Liquide) and isotopically enriched

15N

2

gas (98%, Aldrich Chemical Company Inc.) were used.

However ZnN could also be formed directly albeit in smaller yields from the reaction of a stream of N2 gas and the metal plume without the microwave discharge.

The greatest yield of

ZnN without the microwave discharge came using ammonia (NH3) gas (CIG Industrial Grade Compressed Anhydrous Liquefied) diluted in neon as the reagent gas.

The gas mixture containing the ZnN

ACS Paragon Plus Environment

6

Page 7 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

was deposited on a high purity copper target, cooled to 4.3 K by a continuous-flow liquid helium cryostat (Cryo Industries of America RC110).

A typical deposition would last for 1 hour,

with the laser being continuously rastered across the surface of the zinc target.

After the deposition, the copper target was

inserted into a DM4116 microwave cavity coupled to a Bruker ESP300E ESR spectrometer for analysis.

Analysis of the ESR spectra was conducted by matching a simulated spectrum which was produced with the GEN program,45,46 to the experimental

spectrum.

The

GEN

program

used

an

exact

diagonalization of the spin Hamiltonian with terms included for the zero field splitting interaction, the electron and nuclear Zeeman

interaction

magnetic nucleus.

and

the

hyperfine

interaction

for

each

The various magnetic parameters were varied

to minimize the sum of the squares of the differences between observed

and

particular

simulated

experimental

peaks.

The

parameter

change to

required

produce

a

to

any

simulated

spectrum that differed substantially (around ± 0.5 G) from the experimental spectrum was considered to be the uncertainty for that parameter.

The potential energy surface (PES) of the ZnN radical was constructed through a CASSCF calculation using the ORCA47 program with the uncontracted augmented correlation consistent polarized valence double zeta (aug-cc-pVDZ)48-51

ACS Paragon Plus Environment

7

basis set. The active

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 47

space for the CASSCF calculation contained 7 valence electrons which were distributed across 8 orbitals. The active orbitals involved were 10σ11σ12σ4π4π13σ5π5π. Theoretical determinations of the magnetic parameters were made using a range of ab initio calculations. The ORCA47 program, with the aug-cc-pVDZ48-51 basis set, was used for the B3LYP, B3PW91, MP2 and CASSCF calculations at the optimized geometry determined from the minimum of the PES. A second B3LYP calculation was performed with the triple zeta aug-cc-pVTZ48-51

basis set, to examine the effect of a

larger basis set on theoretical results. A Hartree-Fock singles and doubles configuration interaction (HFSDCI) calculation was performed with the MELDF52 suite of programs and the decontracted aug-cc-pVDZ48-51

basis

set.

The

configuration

integration

included all of the Hartree-Fock (HF) single excitations and the double

excitations

that

had

an

threshold of 1 x 10-7 Hartree.

energy

contribution

over

a

A multi-reference singles and

doubles configuration interaction (MRSDCI) calculation was also performed using the MELDF52 suite of programs and the same basis set

using

100

reference

configurations

from

the

HFSDCI

calculation chosen based on the basis of their CI coefficient. Finally, a Free Atom Comparison Method (FACM) analysis has been performed comparing the experimentally determined values of Aiso and Adip for zinc and nitrogen to the values for the respective free atoms.

RESULTS AND DISCUSSION

ACS Paragon Plus Environment

8

Page 9 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Matrix Isolation Experiments.

Through the reaction of laser

ablated zinc with various N-14 containing reactants, the three sets of triplet peaks displayed in the lower trace of Figure 1 were consistently observed.

Figure 1. The lower trace is the ESR spectrum produced by the Zn14N radical trapped in a neon matrix at 4.3 K. The simulated spectrum

generated

through

the

GEN

program,

using

the

parameters given in Table 2, is shown in the upper trace. The smaller set of accompanying peaks at 5900 G are assigned to a secondary site in the matrix. The line positions of these peaks are presented in Table 1.

The

relative intensity of each set was similar between runs using N atoms, N2 or NH3 as a reactant implying they were associated with the same radical.

ACS Paragon Plus Environment

9

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table

1.

Simulated

Observed

Line

and

Positions

Gauss for the ZnN and

in

67Zn15N

Isotopologues of ZnN Zn14N frequen cy

MIa

Obs.b

Calc.c

1

1883.8( 3)

1883.7 3

0

1890.2( 3)

1890.1 1

-1

1896.6( 3)

1896.4 9

5025.9( 3)

5025.8 6

5029.2( 3)

5029.2 8

5032.8( 3)

5032.7 2

1

5877.2( 3)

5877.3 0

0

5883.7( 3)

5883.6 5

-1

5890.0( 3)

5890.0 4

(MHz) 9702.74 3

9702.76 5

9702.88 3

67Zn15Nd

MIe

MIa

5/2

1/ 2

Obs.b

Calc.c 1746.0 8

ACS Paragon Plus Environment

10

Page 10 of 47

Page 11 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

3/2

1/2

-1/2

-3/2

-5/2

aThe

1/ 2

1755.3( 3)

1/ 2

1800.0( 3)

1/ 2

1808.8( 3)

1/ 2

1854.8( 3)

1/ 2

1863.6( 3)

1/ 2

1910.7( 3)

1/ 2

1919.7( 3)

1/ 2

1967.9( 3)

1/ 2

1976.3( 3)

1/ 2

2026.0( 3)

1/ 2

2035.0( 3)

1755.0 4 1799.6 7 1808.6 7 1854.4 6 1863.4 2 1910.4 7 1919.4 5 1967.6 2 1976.6

2025.8 9 2034.8 8

magnetic quantum number

for the N (I=1) nucleus or the 15N

(I=1/2) nucleus.

bPeak

positions as measured

at the apex. The radical was trapped in a neon matrix at 4.3

ACS Paragon Plus Environment

11

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

K. Uncertainty in position is ± 0.3 G. cPeak

positions

of

a

simulated spectra generated by direct diagonalisation of the spin

Hamiltonian

using

the

magnetic parameters found in Table 2. dMicrowave

frequency

of

9713.955 MHz eThe

magnetic quantum number

for the Zn (I=5/2) nucleus.

A partially overlapped triplet of lower intensity can be seen near 5,900 G which, we attribute to a different trapping site in the matrix.

It is normally assumed that molecules trapped

in rare gas matrices occupy substitutional sites replacing one or more host atoms.

Small interactions between the host and the

trapped species can result in small changes in the spectroscopic properties observed, in this case the D-value, and as the matrix is believed to be crystalline these interactions are specific to a particular site.53

The absorption near 1730 G corresponds

to two spin coupled N atoms, N····N, which has been described by Knight et al.54. The position of the three peaks is consistent with the spectrum generated by a

4-

radical with |D| between

5000 and 10000 MHz. A quartet radical with this |D| value

ACS Paragon Plus Environment

12

Page 12 of 47

Page 13 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 2. The lower trace is the ESR spectrum produced by the Zn15N radical trapped in a neon matrix at 4.3 K. The simulated spectrum

generated

through

the

GEN

program,

using

the

parameters given in Table 2, is shown in the upper trace. The smaller set of accompanying peaks at 5900 G are assigned to a secondary site in the matrix. produces a set of perpendicular (xy1) lines around 1900 G, a set of off angle (oa) lines around 5000 G and another set of perpendicular (xy3) lines around 6000 G, which are all shown in Figure 1. The 1:1:1 triplet splitting pattern observed for the peaks implies a single N-14 nucleus (I = 1) is present in the radical. Repeating the experiment with a N-15 (I = ½) enriched reactant produced the spectrum shown in the bottom trace of Figure

2,

which

exhibits

a

1:1

doublet

splitting

pattern,

confirming the presence of a single nitrogen atom in the radical. The line positions of these peaks are given in Table S1 of the supplementary material. To confirm the presence of zinc atom in the radical, the Zn-67 (I = 5/2, natural abundance 4.1 %) hyperfine lines must be observed.

Given the low natural abundance and the high value

on the nuclear spin, these hyperfine peaks will appear as

ACS Paragon Plus Environment

13

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 47

Figure 3. The lower trace is the ESR spectrum produced by the 67Zn14N

radical trapped in a neon matrix at 4.3 K. The more

intense Zn14N lines, and the N····N interaction lines, have been omitted. The simulated spectrum generated through the GEN program, using the parameters given in Table 2, is shown in the upper trace. satellites to the central multiplet with an intensity of 1/150 of the central multiplet.

The lower trace of Figure 3 shows the

region around the xy1 line for the N-14 labelled ZnN spectrum. For clarity, the xy1 triplet at the center seen from even isotopes of Zn (I=0) and the low field region where the N····N peaks appear have been omitted.

Six sets of triplets with the

same spacing as the xy1 line can be seen and a simulation using the parameters list in Table 2 is given in the upper trace. Table 2. Magnetic Parameters in MHz for the ZnN Radical Isolated in a Neon Matrix at 4.3 K

N

g⊥

g∥

A⊥

A∥

Aisoa

Adipa

1.9998(3)

2.0018 (3)

17.9(20 )

1.5 (20)

11.4(2 0)

7268(8 6.5(13) )

ACS Paragon Plus Environment

14

|D|

Page 15 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

15N

1.9998(3)

2.0018 (3)

25.1(20 )

2.0(20 )

16.1(2 0)

7268(8 9.0(13) )

67Zn

1.9998(3)

2.0018 (3)

156(3)

168(12 )

160(6)

4(5)

aA

⊥ and A∥ were

7268(8 )

used to calculate Aiso and Adip through standard expressions (Ref. 58). The signs on

Aiso and Adip cannot be determined from experiment; hence, they are taken from the theoretical results.

ACS Paragon Plus Environment

15

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The lowest field line of the lowest field triplet is obscured

ACS Paragon Plus Environment

16

Page 16 of 47

Page 17 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

by the N····N peaks.

The ratio of intensity between the central

ACS Paragon Plus Environment

17

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

triplet and the 18 line pattern is about 0.010, compared with

ACS Paragon Plus Environment

18

Page 18 of 47

Page 19 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

the expected value of 0.007. There is some overlap of the sextet

ACS Paragon Plus Environment

19

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 47

Figure 4. The lower trace is the ESR spectrum produced by the 67Zn15N

radical trapped in a neon matrix at 4.3 K. The more

intense Zn15N lines, and the N····N interaction lines, have been omitted. The simulated spectrum generated through the GEN program, using the parameters given in Table 2, is shown in the upper trace. of triplets with other lines so to confirm the assignment experiments were conducted with N-15 labelled N2 and NH3.

This

spectrum is shown in lower trace in Figure 4 with the upper trace being a simulation using the parameters given in Table 2. The substitution of N-15 for N-14 should show a sextet of doublets which is indeed observed.

Again the xy1 doublet at the

center seen from even isotopes of Zn (I=0) and the low field region where the N····N peaks appear have been omitted for clarity.

The ratio of intensity between the central doublet and

the 12 line pattern is also about 0.010, compared with the expected value of 0.007.

The fact that this radical could be

produced from reaction of N2 and Zn confirms no other nuclei are

ACS Paragon Plus Environment

20

Page 21 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

present in the radical and the signal is due to ZnN in a

4-

electronic state. Theoretical Results.

The potential energy surface (PESs) of the

10 lowest lying energy states of the ZnN radical was calculated using the CASSCF procedure with the ORCA program.47 ZnN is shown in Figure 5.

The PES for

The starting orbitals for this

procedure were taken from an MP2 calculation at the equilibrium bond length. Both calculations utilized the aug-cc-pVDZ basis set for Zn and N. The ground state configuration of the ZnN radical

at

re

is

1σ22σ21π43σ24σ25σ22π46σ27σ23π41δ48σ29σ210σ14π2.

Table 3 presents leading contribution of each configuration to the lowest 7 electronic states, the energy of the state relative to

the

ground

state,

dissociation energy.

the

equilibrium

bond

length

and

the

The ground electronic state for ZnN was

found to be 4- with re = 2.079 Å, De = 1.0 kcal/mol and μe = 1.36 D. However around 2.5 Å there is a small barrier on the surface such that the depth of the well at re is 3.5 kcal/mol. The potential energy levels of the dissociated combinations of Zn and N for each state are compared to experimentally determined energy levels of atomic Zn55 and atomic N56 in the supplementary material.

ACS Paragon Plus Environment

21

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 47

Figure 5. The Potential Energy Surface for the ZnN radical. Depicted are the 10 lowest lying states from a CASSCF(7,8) calculation utilizing orbitals taken from an MP2 calculation. Table 3. Bond Lengths (re), Binding Energies (-De), Energy Differences, and Leading Configurations for Different Electronic States of the ZnN Radical Stat e

re (Å)

De (kcal/mol )

Te (eV)a

Configurationb

Leading (%)

4-

2.08

1.0

0

9σ210σ211σ14π14π1

95.1

2-

2.01

48.9

0.55

9σ210σ211σ14π14π1

89.7



2.01

18.0

1.86

9σ210σ24π24π1

88.1



2.15

15.8

1.98

9σ210σ211σ14π2

89.6

2-

1.98

30.0

3.15

9σ210σ211σ14π14π1

89.7

4-

2.64

15.9

3.18

9σ210σ111σ24π14π1

85.6



2.25

69.7

3.56

9σ210σ111σ14π24π1

88.1

i

i

ACS Paragon Plus Environment

22

Page 23 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

aCASSCF

(7,8) was used with the aug-cc-pVDZ basis set (Ref. 48-51)

to calculate theoretical values. bConfigurations

state.

omit core orbitals, taken at re for the particular

cGround

state energy was -254.0286 Hartree. This was set to 0 eV for the calculation of Te.

Using the bond length (re = 2.08 Å ) for ZnN obtained using the CASSCF calculations of the g tensor and |D| value were undertaken using the full MP2 approach and Density Functional Theory (DFT) using the B3PW91 and B3LYP functionals.

These calculations used

the uncontracted aug-cc-pVDZ basis set.

The B3LYP result was

also repeated with the uncontracted aug-cc-pVTZ basis set.

The

results from these calculations are presented in Table 4 with the experimental results. Table

4.

Theoreticala

and

Experimental |D| and g Constants for the ZnN Radical |D|b

g⊥

g∥

B3PW91

8085

1.99982

2.00223 5

B3LYP

7746

1.99997

2.00223 0

B3LYPc

7800

1.99999

2.00222 9

Full MP2

9223

2.00030

2.00223 0

Experime nt

7268(8 )

1.9998( 3)

2.0018( 3)

ACS Paragon Plus Environment

23

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

aAll

calculations

were

Page 24 of 47

performed

using the uncontracted aug-cc-pVDZ basis set for both zinc and nitrogen (Ref.

48-50),

unless

stated

otherwise. b|D|

values given in MHz.

cPerformed

with

the

aug-cc-pVTZ

basis set for both zinc and nitrogen (Ref. 48-51).

Next the hyperfine coupling constants for ZnN were evaluated using

a

range

of

theoretical

approaches.

These

presented in Table 5 with the experimental values.

data

are

The ORCA

program was used to calculate the nitrogen and zinc hyperfine coupling using DFT with the B3PW91 and B3LYP functional, full MP2 and CASSCF.

These calculations used the uncontracted aug-

cc-pVDZ basis set.

A further DFT calculation was undertaken

with the

Table

5.

Theoreticala

Experimental

and

hyperfine

ACS Paragon Plus Environment

24

Page 25 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

coupling constants in MHz for the

67Zn14N

B3PW91

B3LYP

B3LYPb

Full MP2

HFSDCIc,d

MRSDCIc,e

CASSCFf

Experime ntg

aAll

radical Aiso

Adip

Zn

147.4

4.0

N

2.8

-6.3

Zn

149.9

4.1

N

9.8

-6.1

Zn

149.6

4.1

N

9.9

-6.4

Zn

157.3

3.2

N

9.8

-6.6

Zn

179.9

3.6

N

23.9

-8.7

Zn

203.2

4.5

N

9.8

-9.6

Zn

140.2

3.6

N

11.6

-7.0

Zn

160(6 )

4(5)

N

11.4( 20)

6.5(1 3)

calculations

performed

were

using

uncontracted

the

aug-cc-pVDZ

basis set for both zinc and nitrogen

(Ref.

48-51),

unless otherwise stated.

ACS Paragon Plus Environment

25

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

bPerformed

with the aug-cc-

pVTZ basis set for zinc and nitrogen (Ref. 48-51). cThe

CI

calculations

conducted

using

the

were MELDF

suite of programmes. dThe

used

HFSDCI

calculation

all

the

single

excitations provided by the HF configuration plus double excitations contribution

with

an

energy

estimated

to

exceed 1 x 10-7 Hartree. ePerformed

reference

with

100

configurations

chosen on the basis of their coefficients contribution to the HFSDCI wavefunction. All the single excitations from the reference configurations and

the

double

excitations

with an energy contribution estimated to exceed 1 x 10-7 Hartree were retained in the CI calculation.

ACS Paragon Plus Environment

26

Page 26 of 47

Page 27 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

fActive

space of 7,8 using

orbitals generated by the MP2 calculation. gSigns

taken

from

theoretical values.

B3LYP functional and the uncontracted aug-cc-pVTZ basis set.

In

addition, configuration interaction (CI) calculations for the ZnN radical were undertaken with the MELDF program. HFSDCI and MRSDCI calculations were performed using the uncontracted augcc-pVDZ basis set.

As Boldyrev and Simons16 had found a van der

Waals adduct with a bond length of 4.263 Å, we also conducted B3LYP and full MP2 calculations using the ORCA program for this internuclear separation and these results are presented in Table 6 with the experimental hyperfine data determined for This shows our experimental results for

67Zn14N

with a long-bonded van der Waals adduct.

Table

6.

Comparison

of

theoreticala (R = 4.263 Å and measured hyperfine coupling constants for the

B3LYP

67Zn14N

radical

Ato m

Aiso(MH z)

Adip(MH z)

Zn

5.6

0.1

N

10.1

-0.03

ACS Paragon Plus Environment

27

67Zn14N.

are not consistent

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Full MP2

Experime ntb

aAll

performed

Zn

1.57

0.1

N

11.8

0.00

Zn

160(6)

4(5)

N

11.4(1 7)

6.5(11 )

calculations

were

using

Page 28 of 47

the

uncontracted aug-cc-pVDZ basis set for both zinc and nitrogen (Ref. 48-51), unless otherwise stated. bcSigns

taken from theoretical

values.

Formation of ZnN.

The mechanism for the formation of ZnN from

reaction with N atoms or N2 is likely to occur through the association reaction between a zinc atom and a N atom, either produced from the microwave discharge or formed from reaction of N2 molecule with excited atoms in the plume ejected from laser ablation of the zinc metal.

The laser ablation process also

produces quite a large amount of high-energy UV light, as the electrons and atomic ions in the laser ablation plasma recombine and the highly excited atoms fluoresce. The UV light could make excited N2 molecules, which would be far more reactive.

Despite

the very low binding energy, the molecule is stable within a neon matrix at 4.3 K.

For NH3 as the reactant gas, the mechanism

ACS Paragon Plus Environment

28

Page 29 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

of

formation

of

ZnN

could

involve

N

atoms

produced

by

decomposition of the NH3 excited by the plume ejected from laser ablation of the zinc metal, UV photolysis, as discussed above, or N2 molecules which are a known pyrolysis product of NH3.

It

should be noted in all experiments there was a large ESR signal from N atoms and the two spin coupled N atoms, N····N. In our recent work on MgN6 we found acetonitrile was an efficient reactant gas for its production however with zinc only very small amount of ZnN was produced.

Comparison of ZnN with MgN and ZnP.

The experiments reported

here show that the electronic ground state of ZnN is similar radicals with

4-

4-

.

ground states have been studied with

matrix isolation ESR spectroscopy, the MgN radical6 and the radical15.

Two

ZnP

The |D| value for the ZnN molecule (7268 MHz) is

slightly lower than that of MgN (9797 MHz) and far lower than ZnP (29 988 MHz). The difference between ZnN and MgN is likely due to reduced spin-spin interaction from the longer Zn-N bond overruling any increase in spin-orbit interaction due to the heavier zinc nucleus. As the two unpaired π electrons are localized

on

the

nitrogen

atom,

there

will

be

negligible

contributions from the zinc atom to spin-orbit interactions. This also helps explain why the substitution of the phosphorus atom for nitrogen causes a significant increase in |D|, as the larger phosphorus atom has a much greater effect on the spin-

ACS Paragon Plus Environment

29

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 47

orbit interaction of the two π orbitals localized on this atom. The Zn-67 Aiso and Adip values for ZnP (167(30) MHz and 6(4) MHz respectively) are very similar to those of ZnN (159(6) MHz and 4(5) MHz respectively), falling within uncertainty ranges.

As

the zinc hyperfine is related to the charge on the zinc atom this shows both molecules have similar ionic character in the Zn-N and Zn-P bond. The Aiso and Adip values for the N-14 atom follow a similar trend between ZnN and MgN, with very little difference between the two sets of values, they were found to be 11.4(20) MHz and -6.5(13) MHz for ZnN compared to 11.8(12) MHz and -7.8(8) MHz for MgN.

Again, this implies similar ionic

character between the two molecules.

Comparison between theoretical and experimental results.

The

|D| values calculated for ZnN shown in Table 4, overestimated the experimental value by between 6 and 30 %. The best agreement was seen for the B3LYP functional with the uncontracted aug-ccpVDZ basis set and the worst agreement was for the full MP2 calculation.

Increasing the size of the basis set from double

to triple ζ had little effect of the calculated |D| value. A similar result was seen for the MgN radical where again theory overestimated the |D| value by between 6 % and 14 %.

To compare

the calculated g-values it is useful to consider the shifts from ge, defining Δg⊥ = g⊥ - ge and Δg‖ = g‖ - ge.

As expected, Δg‖ is close to

zero and all theoretical methods show good agreement with experiment.

ACS Paragon Plus Environment

30

The

Page 31 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

experimental value for Δg⊥ was -0.0025(3) matched to within given uncertainty to the values predicted by the three DFT methods and only slightly outside the value of -0.002 predicted by the full MP2 approach. For Δg⊥ we would have expected a small positive value given the coupling between the 4-

ground state and the lowest excited 2Πi state as discussed by

Weltner57.

For MgN a positive value for Δg⊥ was observed, and while the DFT and

MP2 methods predicted the correct sign both methods underestimated the magnitude of the shift. For the calculation of the Zn-67 and N-14 hyperfine coupling constants shown in Table 5 the calculated values of the Zn-67 Aiso ranged from 147 to 203 MHz compared with the experimental value of 160(6) MHz and the predicted N-14 Aiso value was around 10 MHz compared with the experimental value of 11(2) MHz. DFT

calculation

agreement

for

Interestingly

with both

the

the the

HFSDCI

B3PW91 Zn-67 and

functional Aiso

MRSDCI

showed Aiso

The

poorest

and

N-14

values.

both

overestimated

the

values for the Zn-67 Aiso coupling constant although the MRSDCI calculation showed better agreement for the N-14 Aiso value than the HFSDCI calculation. All methods employed showed very good agreement with the experimental values for both the Zn-67 Adip and N-14 Adip coupling constants, although, it should be noted that the experimental uncertainty in these values is much larger than for the Aiso values. Engels et al. have noted this phenomena previously57, observing calculated values of Aiso can often differ from measured values by 40-50 %.

Aiso depends on the description

ACS Paragon Plus Environment

31

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 47

of the s electron density at the nucleus whereas Adip is a measure of the spatial distribution of the spin density. As such to calculate

Aiso

wavefunction

reliably well

at

the the

basis

set

nucleus

must

which

describe

is

the

the

reason

uncontracted basis sets were used for these calculations, and as the net spin at the nucleus will be influenced by spin polarization of inner shells electron correlation has to be taken into account.

Bonding and electronic structure.

The Free Atom Comparison

Method (FACM) involves predicting the contribution of the atomic orbitals to the singly occupied molecular orbitals by using the ratio of the hyperfine coupling constants observed for the molecule and the atomic values.

This approach for a



radical

has been described in detail for the BC radical.10 For the ZnN radical there is one unpaired electron in a σ-type molecular orbital, and two unpaired electrons in two π-type molecular orbitals. This gives rise to the following set of equations for the

contribution

of

each

atomic

orbital

to

the

molecular

orbitals: σ = a1 (Zn 4s) + a2 (Zn 4pz) + a5 (N 2s) + a6 (N 2pz), πx = a3 (Zn 4px) + a7 (N 2px ), πy = a4 (Zn 4py) + a8 (N 2py), where

the

separate

contribution wavefunction.

from

variables, each

atomic

a1

through orbital

a8 to

represent the

the

molecular

Symmetry considerations require a3 = a4 and a7 =

ACS Paragon Plus Environment

32

Page 33 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

a8.

However that leaves 6 values to be determined from 4

experimental values so it is assumed that the π-type molecular orbitals are completely composed of the nitrogen 2px and 2py atomic orbitals with one unpaired electron in each, and that the zinc orbitals 4px and 4py are unoccupied.

The one further

unpaired electron is shared across the σ-type orbitals in the radical. By convention the total spin is normalized to one. This simplifies atomic contributions to the following equations: 0.33 = a72 = a82, 0 = a 32 = a 42, a12 = Aiso (molecule)/Aiso (atom), a22 = Adip (molecule)/Adip (atom), a52 = Aiso (molecule)/Aiso (atom), a62 = Adip (molecule)/Adip (atom).

Table 7. FACM and MRSDCI Orbital Population Analysis 67Zn

14N

4s

4pxa

4pya

4pz

2s

2pxb

FACM

0.080 (3)

0

0

0.1( 1)

0.006 (1)

0.33 0.33 0.12 3 3 (2)

1.0( 2)

MRSDCIc

0.146

0.00 0.00 0.04 4 4 5

0.007

0.32 0.32 0.13 5 5 8

0.99 4

aSet

to 0 (refer to text)

bSet

to 0.333 (refer to text)

cPerformed

basis

of

2pyb

2pz

Sum

with 100 reference configurations chosen on the their

coefficients

contribution

ACS Paragon Plus Environment

33

to

the

HFSDCI

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 47

wavefunction.

All the single excitations from the reference

configurations

and

the

double

excitations

with

an

energy

contribution estimated to exceed 1 x 10-7 Hartree were retained in the CI calculation. For the FACM analysis the atomic values for Zn-67 are Aiso = 1992 MHz and Adip = 36 MHz as derived18 from the gas phase level crossing experiments of atomic values

Landman and Novick59 and for N-14 the

are Aiso = 1811 MHz and Adip = 55.52 MHz, obtained

theoretically from Herman and Skillman’s wavefunction.58,

60

The

orbital characters generated by the FACM are presented in Table 7.

Also shown in Table 7 is the population analysis from the

MRSDCI calculation.

The sum of the orbital populations with FACM analysis is unity within the experimental uncertainty.

Comparison of the values

for the MRSDCI Mulliken population analysis show the assumption that the two singly occupied π orbitals would be primarily composed of the N 2p orbitals is justified.

The overall spin contribution

from each atom to the molecular orbitals is matched exactly between the FACM method (Zn67 19 %, N-14 79 %) and the MRSDCI population analysis (Zn-67 20 %, N-14 79 %). However the FACM analysis gives a lower Zn 4s character than the MRSDCI population analysis. Given that the experimental uncertainty in the value for Zn-67 Adip is large, it is not possible to comment on the difference in the Zn 4p character between the two methods.

ACS Paragon Plus Environment

34

Page 35 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The FACM populations indicate that there is similar spin density on the zinc atom, 0.2(1) compared to the spin density on the magnesium atom of MgN 0.22(7). To compare ZnN to ZnP, the FACM for ZnP is redone using the experimental Aiso and Adip values for zinc, which give orbital populations that are more comparable. As such for ZnP, the spin densities the four calculated orbitals are: Zn 4s = 0.08(2), Zn 4pz = 0.02(1), P 3s = 0.0054(6), and P 3pz = 0.23(9). The results for the s orbitals are very similar to ZnN, and there is a shift of density from the pz orbital of the adjacent atom back to the zinc atom in the case of ZnP. Figure 6 shows a molecular orbital diagram for the ground state of ZnN based on the CASSCF(7,8) calculation discussed above. Major contributions are those determined to account for over 10 % of the electron density for that orbital. 4π

orbitals

are

singly

occupied

and

would

While the 10σ and be

expected

to

contribute to the bonding in the molecule the 4π orbitals are almost completely localized on the nitrogen atom and therefore would not make a significant contribution. Which is consistent with the weak bond predicted from the calculations.

This

calculation for ZnN indicated that the 10σ was at a lower energy than the two 4π orbitals. A similar ordering was seen for MgN6 whereas for MgP and ZnP15 the opposite order was predicted. Although in each case all three orbitals are very close to each other in energy.

ACS Paragon Plus Environment

35

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 47

Figure 6. Molecular orbital diagram showing the significant contributions from atomic orbitals to the

4Σ-

ground state of

the ZnN radical. The primary atomic contributions (above 10%) are depicted with solid lines whilst secondary contributions (below 10%) are depicted as dashed lines.

CONCLUSIONS The ZnN radical has been identified for the first time through its capture in matrix isolation. The Zn14N, Zn15N, 67Zn15N

67Zn14N

and

radicals were created through the reaction of NH3 gas or

nitrogen atoms generated by a microwave discharge in N2 gas with a

plume

of

laser

ablated

zinc.

Electron

ACS Paragon Plus Environment

36

spin

resonance

Page 37 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

spectroscopy was used to identify the radical and assign it a set of magnetic parameters. The ESR spectrum of the radical indicated that it exists in a bound

4Σ-

ground state, which ran

contrary to the predictions of Boldeyrev and Simons, whose calculations for ZnN predicted a loosely bound van der Waals type complex. Ab initio calculations were performed at different bond lengths, and the good agreement between the bound state calculations and the experimental data lend support to the theory that the ZnN radical observed in matrix isolation exists in the covalently bound state. The FACM analysis used the magnetic parameters determined through experiment to give an estimation of the spin density distribution for the ZnN radical. Most of the spin density of the single spz electron was found in the pz orbitals of the zinc and nitrogen atoms. A comparison to the related MgN radical showed greater transfer of spin density to the nitrogen atom, which partially explained the changes in magnetic parameters between the two radicals. The MP2 calculation most closely matched the experimental results, and the overall performance of the different ab initio methods was similar to that seen for similar radicals presented previously in literature. A potential energy surface was calculated using the CASSCF method with an active space of 7 electrons in 8 orbitals. The plot shows the ground

4Σ-

state to be weakly bound, but still

bound. The dissociation energies, separation energies, binding energies and bond lengths were shown for each state. The overall

ACS Paragon Plus Environment

37

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 47

potential energy plot was similar to related radicals such as MgN. SUPPORTING INFORMATION The supporting information contains Table S1 which presents the observed and simulated line position for Zn15N and

67Zn14N

and Table S2 which compares the calculated atomic dissociation energy for the various states of ZnN shown in Figure 5. This information is available free of charge via the Internet at http://pubs.acs.org ACKNOWLEDGMENT AJM thanks the Australian Research Council for support of this work under the Small Grants Scheme for the initial funding which supported this work and support from an ARC Research Infrastructure, Equipment and Facilities Grant that was used to purchase the ESR spectrometer at UWA. EK thanks DEST for an Australian Postgraduate Award with stipend and thanks UWA for a Jean Rogerson Postgraduate Scholarship. TSH thanks the Australian Government for a Research Training Program (RTP) Scholarship. Appreciation is expressed to Dr David Feller and Professor E. R. Davidson for use of their MELDF program for calculating the nuclear hyperfine parameters.

ACS Paragon Plus Environment

38

Page 39 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

REFERENCES 1.

King, J. C. C.; Robert J., Modern Nutrition in Health and Disease. 10th Ed.;

Lippincott Williams & Wilkins: New York, 2006.

2.

Vallee, B. L.; Auld, D. S., Zinc Coordination, Function, and Structure of Zinc

Enzymes and Other Proteins. Biochemistry 1990, 29, 5647-5659.

3.

Enthaler, S.; Wu, X. F., In. Zinc catalysis: Applications in organic synthesis; Enthaler,

S. , Wu, X. F. , Eds.; Wiley:Weinheim, 2015.

4.

Adachi, S., Properties of Group-IV, III-V and II-VI Semiconductors. John Wiley &

Sons: Chichester, UK, 2006.

5.

Adachi, S., Properties of Semiconductor Alloys : Group-IV, III-V and II-VI

Semiconductors. John Wiley & Sons: Chichester, U.K., 2009.

6.

Hearne, T. S.; Yates, S.; Wild, D.; McKinley, A., A Matrix Isolation ESR and

Theoretical Study of MgN. J. Chem. Phys. 2017, 147, 044307.

7.

Knight, L. B., Jr.; McKinley, A. J.; Babb, R. M.; Hill, D. W.; Morse, M. D., Electron

Spin Resonance Investigation of Sc2+ in Neon Matrices and Assignment of its Ground Electronic State as X4Σ−. Comparison with Theoretical Calculations. J. Chem. Phys. 1993, 99, 7376-7383.

8.

Knight, L. B., Jr.; Cobranchi, S. T.; Earl, E., Neon Matrix Electron-Spin-Resonance

Investigation of 12C2+ and 13C2+ Confirmation of a 4Σg- Ground Electronic State. J. Chem. Phys. 1988, 88, 7348-7356.

ACS Paragon Plus Environment

39

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

9.

Knight, L. B., Jr.; Herlong, J. O.; Babb, R.; Earl, E.; Hill, D. W.; Arrington, C. A.,

Matrix Isolation Electron Spin Resonance Studies of 28,28Si2+, 28,29Si2+ 29,29Si2+ Ge2+ and 73Ge + 2

Produced by Pulsed Laser Vaporization. Comparison with Theoretical Calculations. J.

Phys. Chem. 1991, 95, 2732-2738.

10.

Knight, L. B., Jr.; Cobranchi, S. T.; Petty, J. T.; Earl, E.; Feller, D.; Davidson, E. R.,

Electron-Spin Resonance Investigations of 11B12C, 11B13C, and 10B12C in Neon, Argon, and Krypton Matrices at 4 K - Comparison with Theoretical Results. J. Chem. Phys. 1989, 90, 690-699.

11.

Knight, L. B., Jr.; Cobranchi, S. T.; Herlong, J. O.; Arrington, C. A., Laser

Vaporization Generation of Al12C, Al13C, Al12C2, and Al13C2 for Rare Gas Matrix Electron Spin Resonance Studies: Experimental–Theoretical Comparisons. J. Chem. Phys. 1990, 92, 5856-5866.

12.

Knight, L. B., Jr.; Winiski, M.; Miller, P.; Arrington, C. A.; Feller, D., Neon Matrix

Electron Spin Resonance and Theoretical Investigations of 10,11BH2, 12,13CH2−, and 12,13CH2+. J. Chem. Phys. 1989, 91, 4468-4476.

13.

Knight, L. B., Jr.; Petty, J. T., Neon Matrix ESR Investigation of 69,71GaAs+

Generated by the Photoionization of Laser Vaporized GaAs(s). J. Chem. Phys. 1988, 88, 481482.

14.

Knight, L. B., Jr.; McKinley, A. J.; Babb, R. M.; Morse, M. D.; Arrington, C. A.,

Laser Vaporization Generation of the SiB and SiAl Radicals for Matrix Isolation Electron Spin Resonance Studies; Comparison with Theoretical Calculations and Assignment of their Electronic Ground States as X4Σ. J. Chem. Phys. 1993, 98, 6749-6757.

ACS Paragon Plus Environment

40

Page 40 of 47

Page 41 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

15.

Fuller, R. O.; Chandler, G. S.; Davis, J. R.; McKinley, A. J., Matrix Isolation ESR and

Theoretical Studies of Metal Phosphides. J. Chem. Phys. 2010, 133, 164311.

16.

Boldyrev, A. I.; Simons, J., Ab Initio Study of the Bonding of Zinc Atoms to First-

and Second-Row Main Group Atoms. Mol. Phys. 1997, 92, 365-379.

17.

Knight, L. B., Jr.; Weltner, W., Jr., Hyperfine Interaction, Chemical Bonding, and

Isotope Effect in ZnH, CdH, and HgH Molecules. J. Chem. Phys. 1971, 55, 2061-2070.

18.

McKinley, A. J.; Karakyriakos, E.; Knight, L. B.; Babb, R.; Williams, A., Matrix-

Isolation ESR Studies of the Various Isotopomers of the CH3Zn and ZnH Radicals: Comparisons with ab Initio Theoretical Calculations. J. Phys. Chem. A 2000, 104, 35283536.

19.

Knight, L. B., Jr.; Mouchet, A.; Beaudry, W. T.; Duncan, M., High-Temperature

Generation and ESR Matrix Isolation Studies of ZnF and CdF in Neon and Argon Matrices. J. Magn. Reson. 1978, 32, 383-390.

20.

Kasai, P. H.; McLeod, D., Electron Spin Resonance Study of Intermetallic Molecules

AgM (M = Group 2 Metal). J. Phys. Chem. 1978, 82, 1554-1559.

21.

Kauffman, J. W.; Hauge, R. H.; Margrave, J. L., Studies of Reactions of Atomic and

Diatomic Chromium, Manganese, Iron, Cobalt, Nickel, Copper, and Zinc with Molecular Water at 15 K. J. Phys. Chem. 1985, 89, 3541-3547.

22.

Zack, L. N.; Sun, M.; Bucchino, M. P.; Clouthier, D. J.; Ziurys, L. M., Gas-Phase

Synthesis and Structure of Monomeric ZnOH: a Model Species for Metalloenzymes and Catalytic Surfaces. J. Phys. Chem. A 2012, 116, 1542.

ACS Paragon Plus Environment

41

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

23.

Brewster, M. A.; Ziurys, L. M., Rotational Spectroscopy of 3D Transition-Metal

Cyanides: Millimeter-Wave Studies of ZnCN (X 2Σ+)+. J. Chem. Phys. 2002, 117, 4853-4860.

24.

Halfen, D. T.; Sun, M.; Harris, B.; Ziurys, L. M., The Microwave and Millimeter

Spectrum of ZnCCH (X̃ 2Σ+): A New Zinc-Containing Free Radical. J. Chem. Phys. 2012, 136, 244310-244310-244310.

25.

Flory, M. A.; Apponi, A. J.; Zack, L. N.; Ziurys, L. M., Activation of Methane by

Zinc: Gas-Phase Synthesis, Structure, and Bonding of HZnCH3. J. Am. Chem. Soc. 2010, 132, 17186-17192.

26.

Chertihin, G. V.; Andrews, L., Reactions of Laser-ablated Zn and Cd Atoms with O2:

Infrared Spectra of ZnO, OZnO, CdO, and OCdO in Solid Argon. J. Chem. Phys. 1997, 106, 3457-3465.

27.

Greene, T. M.; Brown, W.; Andrews, L.; Downs, A. J.; Chertihin, G. V.; Runeberg,

N.; Pyykko, P., Matrix Infrared Spectroscopic And Ab-Initio Studies Of ZnH2, CdH2, And Related Metal Hydride Species. J. Phys. Chem. 1995, 99, 7925-7934.

28.

Wang, X.; Andrews, L., Zinc and Cadmium Dihydroxide Molecules:  Matrix Infrared

Spectra and Theoretical Calculations. J. Phys. Chem. A 2005, 109, 3849-3857.

29.

Chang, S.-C.; Hauge, R. H.; Kafafi, Z. H.; Margrave, J. L.; Billups, W. E., Isolation

and Characterization of ZnCH2 by Fourier Transform I.R. Matrix Isolation Spectroscopy and its Photolytic Rearrangement to HZnCH. J. Chem. Soc., Chem. Commun. 1987, 1682-1684.

ACS Paragon Plus Environment

42

Page 42 of 47

Page 43 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

30.

Greene, T. M.; Andrews, L.; Downs, A. J., The Reaction of Zinc, Cadmium, and

Mercury Atoms With Methane - Infrared-Spectra of the Matrix-Isolated Methylmetal Hydrides. J. Am. Chem. Soc. 1995, 117, 8180-8187.

31.

Billups, W. E.; Konarski, M. M.; Hauge, R. H.; Margrave, J. L., Activation of

Methane with Photoexcited Metal Atoms. J. Am. Chem. Soc. 1980, 102, 7393-7394.

32.

McDonald, S. A.; Andrews, L., Matrix Reactions of Zn, Cd and Mg Atoms With NO2.

Infrared spectra of M+NO2− species. J. Mol. Spectrosc. 1980, 82, 455-458.

33.

Ault, B. S.; Andrews, L., Ultraviolet Absorption Spectra of Zn2 and Cd2 in Solid

Argon and Krypton at 10 K. J. Mol. Spectrosc. 1977, 65, 102-108.

34.

Miller, J. C.; Andrews, L., Absorption Spectra of Heteronuclear Group II Metal

Diatomic Molecules in Solid Argon at 10 K. J. Chem. Phys. 1978, 69, 3034-3038.

35.

Henchy, C.; Kilmartin, U.; McCaffrey, J. G., Vacuum Ultraviolet Spectroscopy and

Photochemistry of Zinc Dihydride and Related Molecules in Low-Temperature Matrices. J. Phys. Chem. A 2013, 117, 9168-9178.

36.

Behm, J. M.; Blume, T.; Morse, M. D., Interaction of an Aluminum Atom with a

Closed Subshell Metal Atom: Spectroscopic Analysis of AlZn. J. Chem. Phys. 1994, 101, 5454-5463.

37.

Povey, I. M.; Bezant, A. J.; Corlett, G. K.; Ellis, A. M., Spectroscopic Investigation of

Zinc-Containing Organometallic Radicals Prepared Using a Pulsed Electrical Discharge Nozzle. J. Phys. Chem. 1994, 98, 10427-10431.

ACS Paragon Plus Environment

43

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

38.

Robles, E. S. J.; Ellis, A. M.; Miller, T. A., Electronic Spectroscopy of Jet-cooled

Half-sandwich Organometallic Free Radicals. 2. Laser-induced Fluorescence Study of the Pyrrolyl Complexes of Zinc and Cadmium. J. Phys. Chem. 1992, 96, 3258-3265.

39.

Robles, E. S. J.; Ellis, A. M.; Miller, T. A., Electronic Spectroscopy of Jet-cooled

Half-sandwich Organometallic Free Radicals: Laser-induced Fluorescence Study of the Monomethylcyclopentadienyl Complexes of Zinc and Cadmium. J. Chem. Soc., Faraday Trans. 1992, 88, 1927-1941.

40.

Robles, E. S. J.; Ellis, A. M.; Miller, T. A., Electronic Spectroscopy of Jet-cooled

Half-sandwich Organometallic Ffree Radicals. 1. Laser-induced Fluorescence Study of the Cyclopentadienyl Complexes of Zinc and Cadmium. J. Phys. Chem.1992, 96, 3247-3258.

41.

Cernicharo, J.; Guelin, M., Metals in IRC+10216 - Detection of NaCl, AlCl, and KCl,

and Tentative Detection of AlF. Astron. Astrophys. 1987, 183, L10-L12.

42.

Neufeld, D. A.; Zmuidzinas, J.; Schilke, P.; Phillips, T. G., Discovery of Interstellar

Hydrogen Fluoride. Astrophys. J. Lett. 1997, 488, L141-L144.

43.

Neufeld, D. A.; Schilke, P.; Menten, K. M.; Wolfire, M. G.; Black, J. H.; Schuller, F.;

Muller, H.; Thorwirth, S.; Gusten, R.; Philipp, S., Discovery of Interstellar CF+. Astron. Astrophys. 2006, 454, L37-L40.

44.

Karakyriakos, E.; Davis, J. R.; Wilson, C. J.; Yates, S. A.; McKinley, A. J., Neon and

Argon Matrix ESR and Theoretical Studies of the 12CH3Cd, 12CD3Cd, 13CH3Cd, 12CH3111Cd, and 12CH3113Cd radicals. J. Chem. Phys. 1999, 110, 3398-3410.

ACS Paragon Plus Environment

44

Page 44 of 47

Page 45 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

45.

Knight, L. B., Jr.; Cobranchi, S. T.; Earl, E., Neon Matrix ESR Investigation of 12C+2

and 13C+2 : Confirmation of a 4Σg− Ground Electronic State. J. Chem. Phys. 1988, 88, 73487356.

46.

Knight, L. B., Jr.; Steadman, J.; Miller, P. K.; Cleveland, J. A., ESR Investigation of

the Cation Radicals 14N213CO+, 15N212CO+, and 15N213CO+: The Trapping of Ion–Neutral Reaction Products in Neon Matrices at 4 K. J. Chem. Phys. 1988, 88, 2226-2234.

47.

Neese, F., The ORCA program system. Wiley Interdiscip. Rev.: Comput. Mol. Sci.

2012, 2, 73-78.

48.

Dunning Jr, T. H., Gaussian Basis Sets for Use in Correlated Molecular Calculations.

I. The Atoms Boron Through Neon and Hydrogen. J. Chem. Phys. 1989, 90, 1007-1023.

49.

Woon, D. E.; Dunning Jr, T. H., Gaussian Basis Sets for Use in Correlated Molecular

Calculations. III. The Atoms Aluminum Through Argon. J. Chem. Phys. 1993, 98, 13581371.

50.

Wilson, A. K.; Woon, D. E.; Peterson, K. A.; Dunning Jr, T. H., Gaussian Basis Sets

for Use in Correlated Molecular Calculations. IX. The Atoms Gallium Through Krypton. J. Chem. Phys. 1999, 110, 7667-7676.

51.

Prascher, B. P.; Woon, D. E.; Peterson, K. A.; Dunning, T. H., Jr.; Wilson, A. K.,

Gaussian Basis Sets for Use in Correlated Molecular Calculations. VII - Valence, CoreValence, and Scalar Relativistic Basis Sets for Li, Be, Ba, and Mg. Theor. Chem. Acc. 2011, 128, 69-82.

ACS Paragon Plus Environment

45

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

52.

Brom, J. M.; Weltner, W., Jr., ESR Spectrum and Structure of the MgOH Radical. J.

Chem. Phys. 1973, 58, 5322-5330.

53.

Dunkin, I. R., Matrix-isolation Techniques: A Practical Approach. Oxford University

Press: Oxford, 1998; p19.

54.

Knight, L. B., Jr.; Bell, B. A.; Cobranchi, D. P.; Davidson, E. R., Electron Spin

Resonance and Theoretical Studies of the 14N⋅⋅⋅⋅14N and 15N⋅⋅⋅⋅15N Spin-Pair Radicals in Neon Matrices: The Effects of Mixing Among the 1Σg+, 3Σu+, 5Σg+, and 7Σu+ Electronic States. J. Chem. Phys. 1999, 111, 3145-3154.

55.

Dag, G.; Ulf, L., Accurately Measured Wavelengths of Zn I and Zn II Lines of

Astrophysical Interest. Phys. Scr. 2000, 61, 652.

56.

Moore, C. E., Natl. Stand. Ref. Data Ser., Natl. Bur. Stand. (U.S.) 1975, 3.

57.

Engels, B.; Peyerimhoff, S. D.; Davidson, E. R. Calculation of hyperfine coupling

constants. An ab ititio MRS-CI study for nitrogen to analyse the effects of basis sets and CI parameters Mol. Phys. 1987, 62, 109-127.

58.

Weltner, W., Jr., Magnetic Atoms and Molecules. Dover Publications: New York,

1983.

59.

Landman, A.; Novick, R., Level Crossing Determination of τ(1P1) and gj(3P1) in Zinc

and the hfs of Zn65 and Zn67. Phys. Rev. 1964, 134, A56-A65.

60.

Herman, F.; Skillman, S., Atomic Structure Calculations. Prentice-Hall Inc.: New

Jersey, 1963.

ACS Paragon Plus Environment

46

Page 46 of 47

Page 47 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

TOC GRAPHIC

ACS Paragon Plus Environment

47