Document [original]

Adamantane Very Important Paper

Infrared Spectrum of the Adamantane+–Water Cation:Hydration-

Induced C@HBond Activation and Free Internal Water Rotation

Martin Andreas Robert George,Marko Fçrstel, and Otto Dopfer*

Abstract: Diamondoid cations are reactive intermediates in

their functionalization reactions in polar solution. Hydration is

predicted to strongly activate their C@Hbonds in initial proton

abstraction reactions.Tostudy the effects of microhydration on

the properties of diamondoid cations,wecharacterize herein

the prototypical monohydrated adamantane cation

(C10H16+–H2O, Ad+–W) in its ground electronic state by

infrared photodissociation spectroscopyinthe CH and OH

stretch ranges and dispersion-corrected density functional

theory (DFT) calculations.The water (W) ligand binds to the

acidic CH group of Jahn–Teller distorted Ad+via astrong

CH···O ionic H-bond supported by charge–dipole forces.

Although Wfurther enhances the acidity of this CH group

along with aproton shift toward the solvent, the proton remains

with Ad+in the monohydrate.Weinfer essentially free internal

Wrotation from rotational fine structure of the n3band of W,

resulting from weak angular anisotropyofthe Ad+–W

potential.

Introduction

Adamantane (C10H16,Ad) is the parent molecule of

diamondoids,which are nanometer-sized H-passivated nano-

diamonds.[1] These rigid and stress-free cycloalkanes and their

derivatives have well-defined stable structures and thus are of

interest in avariety of disciplines,with (potential) applica-

tions ranging from materials and polymer sciences,molecular

electronics,medical sciences,and chemical synthesis to

astrochemistry.[2] Specifically,radical cations of diamondoids

are intermediates in the their functionalization reactions in

polar solvents.[2d,3] Quantum chemical calculations predict

asubstantial activation and eventually rupture of the most

acidic C@Hbond in these radical cations caused by solvation

with afew polar solvent molecules,such as water (H2O, W) or

acetonitrile.[2d,3] This appears to be the first step required for

Hsubstitution via aradical cation mechanism.

To understand the involved reaction mechanism at the

molecular level, detailed knowledge of the interaction

between the diamondoid radical cation and the solvent

molecules is required. To this end, we characterize the

interaction between Ad+,the parent cation of the diamondoid

family,and Wligands by 1) infrared photodissociation

(IRPD) spectroscopy of mass-selected (Ad–Wn)+clusters

generated in amolecular beam and 2) dispersion-corrected

density functional theory (DFT) calculations.Herein, we

report the results obtained for monohydrated Ad+–W.

Significantly,these results represent the first experimental

(and in particular spectroscopic) characterization of any

diamondoid cation interacting with apolar solvent.

Thescarce knowledge of Ad+comes from photoelectron

and fragmentation spectroscopy,[4] IR spectroscopy,[5] and

calculations.[2d,3a,5,6] Neutral Ad has astructure with Td

symmetry,four equivalent CH groups,and six equivalent

CH2groups (Figure 1).[7] Ionization from the fully occupied

triply degenerate (7t2)6HOMO leads to the Jahn–Teller

distorted 2A1cation ground state with (12e)4(12a1)1config-

uration and C3vsymmetry.[5,6] Removal of the bonding

electron leads to apronounced elongation of one of the

C@Hbonds along the C3axis.This predicted activation has

experimentally been probed in IRPD spectra of Ad+–Henand

Ad+–N2clusters via its unusually low C@Hstretch frequency

at 2600 cm@1.[5] Similar C@Hbond activation has more

recently been observed for ionization of linear alkanes and

their hydrated clusters.[8] Theanalysis of the IRPD spectrum

of monohydrated Ad+–W elucidates the interaction potential

between Ad+and Wwith respect to the binding site,the

strength and anisotropy of the interaction, and the magnitude

of hydration-induced C@Hbond activation.

Results and Discussion

TheIRPD spectrum of mass-selected Ad+–W clusters is

measured in the Wloss channel in atandem quadrupole mass

spectrometer coupled to an electron ionisation source using

an optical parametric oscillator laser system,[9] asetup

previously used to record IR spectra of hydrocarbon cation

clusters.[5,10] Theinvestigated range (2400–3900 cm@1)covers

the informative OH and CH stretch modes (nOH,nCH(2)).

Quantum chemical calculations are performed at the dis-

persion-corrected B3LYP-D3/cc-pVTZ level to elucidate the

energetic,structural, electronic,and vibrational properties of

Ad, Ad+,and its Ad+–W clusters.[11] Additional calculations at

other DFT levels yield essentially the same results.Relative

energies (Ee)and binding energies (De)are corrected for

harmonic zero-point vibrational energies to derive E0and D0.

Thecomputed structures and scaled harmonic IR spectra

of W, Ad, and Ad+shown in Figure 1, Figure 2, and Figure S1

[*] M. A. R. George, Dr.M.Fçrstel, Prof. Dr.O.Dopfer

Institut ffrOptik und Atomare Physik, Technische Universit-tBerlin

Hardenbergstrasse 36, 10623 Berlin (Germany)

E-mail:d[email protected]

Supportinginformation and the ORCID identification number(s) for

the author(s) of this article can be found under:

https://doi.org/10.1002/anie.202003637.

T2020 The Authors. Published by Wiley-VCH Verlag GmbH &Co.

KGaA. This is an open access article under the terms of the Creative

Commons Attribution Non-Commercial License, which permits use,

distribution and reproduction in any medium, provided the original

work is properly cited, and is not used for commercial purposes.

ngewandte

Chemie

Research Articles

How to cite: Angew.Chem. Int. Ed. 2020,59,12098–12104

International Edition: doi.org/10.1002/anie.202003637

German Edition: doi.org/10.1002/ange.202003637

12098 T2020 The Authors. Published by Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim Angew.Chem. Int. Ed. 2020,59,12098 –12104

in the Supporting Information agree well with available

computational and experimental data,[2d,3a,5,6,12] indicating

that the B3LYP-D3/cc-pVTZ level describes the monomer

units well. Ionization of Ad (1A1,Td)into its ground electronic

state results from removal of abonding electron from the t2

orbital, giving rise to Jahn–Teller distortion in the resulting

2A1state with C3vsymmetry.[5] According to the shape of the a1

orbital (Figure S2), the major structural impact of ionization

are an elongation of the three C@Cbonds parallel to the C3

axis by 72 mc(the other C@Cbonds contract by 27 mcand

16 mc)and adrastic elongation of the C@Hbond on the C3

axis by 30 mc.This C@Hbond located at the top of Ad+

becomes very acidic and carries the largest positive partial

charge of all protons (354 me,Figure S3). As aresult, the

corresponding nCHtfrequencyisrather low (2606 cm@1), in

agreement with experiment (Figure 2).[5]

TheJahn–Teller distorted Ad+cation offers three attrac-

tive binding sites for W, which can be attached to the

top (Ia/b), the bottom (II), and the side (III)ofthe distorted

Ad+tetrahedron (Figure 1). All structures have afavourable

charge–dipole configuration, with the Oatom of Wpointing

toward the Ad+cation. Because the ionization energy of Ad is

much lower than that of W(9.25 vs.12.6 eV)[12] and the

computed proton affinity of the adamantyl radical (C10H15)is

much higher than that of W(PA =868 vs.691 kJmol@1),[12]

neither charge nor proton transfer occurs upon monohydra-

tion, justifying the notation of Ad+–W in the ground

electronic state.The computed linear IR spectra of all Ad+–

Wisomers in the nOH/CH(2) range are compared in Figure 2to

that of Wand Ad+(Table 1and Figure S1). In the Ad+–W(Ia)

global minimum, Wforms astrong and nearly linear CH···O

ionic H-bond to the acidic CH group at the top of Ad+,with

ahigh binding energy of D0=45.6 kJmol@1and ashort

intermolecular distance of RCH···O =1.703 c.The Ia structure

is similar to the one calculated previously at the lower B3LYP/

6-31G* level.[3a] Wisnot directly aligned along the C3

rotational axis,with aCH···O bond angle of qCHO =170.288

and an angle of qC2 =46.288between the C2axis (baxis) of W

and the intermolecular axis,indicating that the CH group

binds to one of the two lone pairs of W. Upon hydration, the

acidic C@Hdonor bond is further elongated by 49 mc

compared to Ad+(rCH =1.174 c), leading to substantial

additional activation. As aresult, its bound CH stretch

frequency nCHtshifts down from 2606 to 2033 cm@1(outside

Figure 1. Calculated equilibrium structures (in [b]) of W, Ad (Td), Ad+(C3v), Ad+–W(Ia/b-III)intheir ground electronicstate (B3LYP-D3/cc-pVTZ).

Further details are available in Figure S1. Red O, gray H, white H.

ngewandte

Chemie

Research Articles

12099Angew.Chem. Int.Ed. 2020,59,12098 –12104 T2020 The Authors. Published by Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim www.angewandte.org

15213773, 2020, 29, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/anie.202003637 by Technische Universitaet Berlin, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

the scanning range) along with adrastic enhancement in the

IR intensity.This massive redshift of 573 cm@1(or 22%)

illustrates that IR spectroscopy is asensitive probe for

chemical bond activation. Thesubstantial charge transfer of

124 me from Ad+to Wupon formation of the CH···O H-bond

(Figure S3) is also visible in the HOMO orbital (Figure S2). It

results in an elongation of the O@Hbonds (DrOH =4mc),

with aconcomitant redshift in the free OH stretch frequencies

(@Dn1/3 =45/43 cm@1)and astrong IR enhancement, as is

typical for cation–W dimers.[9b,10h,i,13] Apart from the acidic

CH group,monohydration has only aminor impact on the

structure of the remaining Ad+cage (DrCC <14 mc,DrCH ,

4mc). As aresult, the remaining three CH and twelve CH2

stretch frequencies do not change much (DnCH(2) ,10 cm@1),

with the notable exception of the three symmetric CH2

stretches at the bottom of the cage, nCH2(s)b.These are

blueshifted by 27–40 cm@1and strongly suppressed in IR

intensity by monohydration

(band A), in line with the computed C@Hbond contraction

(4 mc). Band C, with aconvoluted maximum at 3000 cm@1,is

composed of the three high-frequencyantisymmetric CH2

stretches of the top of Ad+,nCH2(a)t,slightly redshifted from

those of Ad+.The most intense band Binthe free CH(2)

stretch range peaking at 2956 cm@1contains the three bottom

CH stretches, nCHb,and the remaining three nCH2(s)tand three

nCH2(a)bmodes (strongly coupled), whereby most of the

intensity is coming from two modes at 2956 and 2957 cm@1,

nearly unshifted from those of Ad+(@5cm

@1).

Internal rotation of the Ad+cage of Ia by approximately

6088around the C3axis results in asecond and essentially

isoenergetic isomer Ib (E0=0.26 kJmol@1), with very similar

structural, energetic,and vibrational properties as Ia (Fig-

ure 1, Figure 2, Figure S1, Table 1). Its CH···O H-bond is

characterized by D0=45.3 kJmol@1,RCH···O =1.71 c,qCHO =

170.388,qC2 =46.088,and Dq=123 me.The activation of the

acidic C@Hbond is similar as in Ia (rCH =1.173 c,nCH =

2040 cm@1), and also the properties of Ware comparable.

Thepotential for internal rotation of Ad+around its C3

axis has three equivalent global minima Ia (E0=Ee=0) at 088,

12088,and 24088,and three local minima Ib at approximately

6088,18088,and 30088(E0=0.26 kJmol@1,Ee=0.03 kJmol@1),

which are separated by six low-energy transition states TS1 at

approximately 3088,9088,15088,21088,27088,and 33088with

abarrier of Vb=0.40 kJmol@1above Ia (qCHO =168.088,qC2 =

46.288,RCH···O =1.707 c,rCH =1.173 c,Figure S4). There are

further low-energy TS2 at Vb=2.68 kJmol@1(qCHO =177.688,

RCH···O =1.735 c,rCH =1.163 c,Figure S4), which connect Ia

and Ib minima separated by 18088via aflipping motion of W,

whereby the CH group changes the bonding from one to the

other lone pair of W. In contrast to Ia/band TS1,the baxis of

Wis(nearly) parallel to the intermolecular axis of Ad+(qC2 =

1.588), so that the barrier for internal Wrotation around this

axis approaches zero (Vb<10 cm@1), while those for Ia/bare

rather high (Vb=27 kJmol@1,Figure S5) because of steric

hindrance with the Ad+cage and the rupture of the H-bond to

the lone pair. In conclusion, the Ad+–W potential at the top of

Ad+is rather flat, and the calculations predict four non-

equivalent low-energy stationary points (Ia/b,TS1/2), with

essentially the same energy and IR spectra in the nOH/CH(2)

range (Figures 2and S6), so that they cannot by distinguished

by the measured IRPD spectrum.

In the low-energy isomers II and III (E0=3.50 and

4.79 kJmol@1), Wdoes not bind to Ad+via asingle,strong,

and nearly linear CH···O bond to the acidic CH proton but

forms instead three nonlinear CH+···O H-bonds to three

different CH2groups at either the side or the bottom of the

Ad+cage.Although the individual CH···O contacts in II and

III are much longer than those in Ia and Ib (2.26–2.61 vs.ca.

1.71 c)and charge transfer from Ad+to Wislargely reduced

(from 124 to 23 and 9me), their binding energies of D0=42.1

and 40.8 kJmol@1remain comparable at the B3LYP-D3/cc-

pVTZ level. TheAd

+···W bonding in II and III is essentially

based on charge–dipole forces and has only little contribution

from H-bonding.Asaresult of the minor charge transfer, the

O@Hbonds of Ware much less affected (DrOH =1mc),

leading to smaller red shifts (@Dn1/3 =12/20 and 9/19 cm@1)

and lower IR activity (Figure 2, Table 1). While the IR spectra

of Ia and Ib are quite similar in both the investigated CH and

OH stretch ranges,those of II and III are rather different.

Significantly,the acidic free CH stretch of Ad+at 2606 cm@1

remains for II and III in this frequencyrange (nCHt=2642 and

Figure 2. IRPD spectra of Ad+–W and Ad+–He2[5] compared to linear

and scaled harmonic IR absorption spectra of Ad+,W,and

Ad+–W(Ia/b-III)calculated at the B3LYP-D3/cc-pvTZ level (scaling

factor 0.963). The positions of the transition observed in the IRPD

spectrum of Ad+–W (A–E) and their vibrational assignmentare listed

in Table 1. Band Econsists of three sharp Qbranches owing to internal

Wrotation.

ngewandte

Chemie

Research Articles

12100 www.angewandte.org T2020 The Authors. Published by Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim Angew.Chem. Int. Ed. 2020,59,12098 –12104

2607 cm@1)with high IR activity,because this bond is little

affected by hydration at the side and bottom of Ad+(DrCH ,

5mc). On the other hand, formation of the three CH···O

contacts in II shifts the corresponding bonded nCH2 mode

down to 2776 cm@1and its high IR activity makes this band the

most intense transition (A1). In the IR spectra of both II and

III,further nCH2 frequencies shift down to well below

2900 cm@1(A1-A3) with enhanced IR activity generating

acharacteristic fingerprint. Bands Band Cremain rather

unchanged from those of Ia/b.The transition states for

interconverting II/III into Ia/bare quite high (16.1/

14.7 kJmol@1)sothat II/III may be trapped in their deep

minima in the supersonic expansion.

Thestructures,energies,and IR spectra computed at the

other DFT levels yield the same qualitative picture (Table S2,

Figure S7), with very similar energies for the nearly isoener-

getic Ia/bminima (DE0,0.5 kJmol@1), asubstantial gap to II

and III (4–13 kJmol@1), and the same IR spectral pattern.

In Figure 2, we show the IRPD spectrum of Ad+–W to the

linear IR absorption spectra computed for Ad+,W,and the

four Ad+–W isomers Ia,Ib,II,and III.For comparison, we

also include the IRPD spectrum of Ad+–He2reported

previously,which provides aclose approximation to the

spectrum of bare Ad+due to the weak bonding of He (D0<

1kJmol@1).[5] TheIRPD spectrum of Ad+–W reveals five

vibrational transitions A–E, and their positions,widths,and

vibrational and isomer assignments are listed in Table 1. The

bands A–C occurring in the CH(2) stretch range at 2875, 2942,

and 2976 cm@1with widths of 13–23 cm@1show only minor

shifts from the corresponding bands of Ad+–He2at 2868/2883

(A1/A2), 2941/2954 (B1/B2), and 2981 (C) cm@1(Dn <6cm

for the band centers), although the latter spectrum shows

higher spectral resolution owing to colder ions resulting from

the smaller D0value.The stronger interaction with Wchanges

the relative intensities of these

bands.Inaddition, the intense band

at nCHt=2600 cm@1assigned to the

acidic CH group of Ad+is shifted

away from the investigated spectral

range,providing the first strong

experimental evidence for the for-

mation of aCH···O H-bond in Ad+–

W(isomer I). Finally,two new tran-

sitions Dand Eappear in the Ad+–

Wspectrum at 3625 and 3717 cm@1,

which are readily assigned to the

free OH stretch modes of W(n1/3).

Themagnitude of their redshifts

from bare W(@Dn1/3 =32/39 cm@1)

and their relative intensities are

indicative of astrongly bonded

cation–dipole structure.While the

unresolved n1band (D) has only

aslightly smaller width than the

partly unresolved nCH bands (13 vs.

20–25 cm@1), the n3band (E) shows

rotational substructure with narrow

equidistant Qbranches at 3703,

3731, and 3759 cm@1,with awidth

of 3–4 cm@1and aspacing of around 28 cm@1,indicative of

internal rotation of the Wligand. Interestingly,the two bands

marked by asterisks in the Ad+–He2spectrum (Figure 2)

assigned to overtone and/or combination bands[5] disappear in

the Ad+–W spectrum because of the significant impact of W

on the Ad+cage.

Comparison of the IRPD spectrum of Ad+–W to the

calculated ones unambiguously shows that Wbinds to the

acidic CH group by forming aCH···O ionic H-bond, with

convincing agreement for isomers Ia/bconcerning both

positions and relative intensities of all transitions observed

(with maximum, mean, and summed deviations of 26, 15, and

14 cm@1). Moreover,the shifts and intensity changes of the

nCH(2) bands A–C predicted for He!Wsubstitution are well

reproduced by experiment. Thematch between predicted and

measured nCH(2) bands indicates that anharmonic couplings

between CH stretch fundamentals and CH bend overtones

are not important for this system (at least at the achieved

spectral resolution). On the other hand, the most intense

transitions of the higher energy isomers II and III between

2600 and 2700 cm@1are absent in the measured spectrum and

also their predicted Dn1/3 redshifts are much smaller than the

observed ones.Inaddition, the computed n1frequencies of

Ia/band II/III differ by as much as about 35 cm@1and no such

splitting is observed experimentally.Consequently,the IRPD

spectrum of Ad+–W is solely assigned to Ia/b,while II and III

are below the detection limit (consistent with their lower

stability,population estimated as <3%of Ia/bby considering

the achieved signal-to-noise ratio and the computed IR

intensities). Because Ia and Ib have very similar energies and

IR spectra, the IRPD spectrum is of insufficient resolution to

exclude one or the other.Indeed, the computations predict

low barriers for their interconversion (Ia $Ib)sothat the

zero-point vibrational level lies probably above these barriers

Table 1: Computed vibrationalfrequencies(in cm@1,B3LYP-D3/cc-pVTZ) of Ad+,W,and Ad+–W(Ia/b)

and experimental values of Ad+He2and Ad+–W.[a]

Mode[b] Ad+(C3v)W(C2v)Ad

+–W(Ia)Ad

+–W(Ib)Ad

+–He2exp[c] Ad+–W exp

nCHt2606 (185,a1)2033 (3016) 2040 (2994) 2600

nCH2(s)b2852 (48,e) 2890 (6) 2889 (7) 2868 A1 2875 (23) A

nCH2(s)b2852 (48,e) 2892 (5) 2892 (5) 2868 A1 2875 (23) A

nCH2(s)b2880 (7,a1)2907 (0.2) 2906 (0.6) 2883 A2 2875 (23) A

nCHb2939 (5,e) 2937 (3) 2935 (2) 2941 B1

nCHb2939 (5,e) 2938 (3) 2938 (4) 2941 B1

nCHb2948 (0.1,a1)2944 (3) 2944 (4)

nCH2 2953 (2,e) 2949 (1) 2949 (1)

nCH2 2953 (2,e) 2950 (0.6) 2950 (1)

nCH2 2955 (2,a1)2951 (2) 2952 (5)

nCH2 2961 (16,e) 2956 (37) 2956 (35) 2954 B2 2942 (20) B

nCH2 2961 (16,e) 2957 (35) 2957 (32) 2954 B2 2942 (20) B

nCH2 2968 (1,a1)2962 (1) 2962 (2)

nCH2(a)t3005 (0,a2)2995 (4) 2994 (4)

nCH2(a)t3010 (10,e) 3000 (12) 2998 (15) 2981 C2976 (13) C

nCH2(a)t3010 (10,e) 3002 (15) 3005 (12) 2981 C2976 (13) C

n13658 (3,a1)3613 (206) 3614 (207) 3625 (13) D

n33754 (41,b1)3711 (111) 3711 (111) 3717[d] E

[a] Forcalculated harmonic frequencies(scaled by 0.963), IR intensities (in kmmol@1)and symmetry

species are listed in parentheses. Forexperimental values the width of the band is given in parentheses.

[b] The six nCH2 modes are amixture of nCH2(s)tand nCH2(a) for Ia/b.[c] Ref. [5].[d] Band origin n0.

ngewandte

Chemie

Research Articles

12101Angew.Chem. Int.Ed. 2020,59,12098 –12104 T2020 The Authors. Published by Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim www.angewandte.org

leading to avibrationally averaged structure between the two

nonequivalent triply degenerate minima, denoted structure I.

Thebinding energy of Ad+–W(Ia)computed at the highest

level (B3LYP-D3/aug-cc-pVTZ) is D0=35.4 kJmol@1or

2960 cm@1(Table S2), indicating that single-photon IRPD in

the CH and OH stretch ranges is feasible for cold cluster ions

in the ground or low-energy vibrationally excited states.

Interestingly,the n3band (E) of Ad+–W shows rotational

fine structure.Three narrow Qbranches spaced by 28 cm@1

indicate an effective internal rotational constant of Aeff =

14 cm@1(Figure S8). Therotational constants of W(AW=

27.88, BW=14.52, and CW=9.28 cm@1)[12] suggest to assign

this spectral feature to nearly free internal rotation of the W

subunit around its baxis because BW&Aeff.The observed Aeff

value is slightly smaller than BWdue to the presence of the

heavy Ad+unit and/or asmall nonvanishing barrier.The

selection rules for the observed Qbranches of the perpen-

dicular n3component are DJ=0and Dk=:1, where Jis the

total rotational quantum number of Ad+–Wand kthe internal

rotational quantum number for Wrotation. According to the

Pauli principle for fermions,even and odd krotational levels

are combined with symmetric (triplet, I=1, MI=0and :1)

and antisymmetric (singlet, I=0, MI=0) nuclear spin wave

functions for exchange of the two equivalent protons (nuclear

spin i=1/2), with nuclear spin statistical weights of 3:1,

respectively.Asaresult, supersonic cooling freezes the klevel

populations down to k=0/1 for k=even/odd levels with

aweight of 1/3. Thethree remaining PQand RQbranches (in

the notation DkDJwith DJ=0) expected for low temperature

are then those originating from k=0(1

0, weak) and k=1

1, 2

1, strong), and these are indeed the detected Q

branches (Figure S8, Table S3). From the absence of any

transition originating from k=2, an upper limit for the

(rotational) temperature is estimated as T=70 K. The n3band

origin (n0)occurs at the 0

0subband, which is forbidden for

apurely perpendicular transition but becomes weakly al-

lowed for ahybrid and indeed we observe aweak signal where

the parallel component is expected. The n1band does not

exhibit any Qbranches because the selection rules for this

mostly parallel transition are DJ=:1and Dk=0, and the

resulting Pand Rbranches of the overall end-over-end

rotation of Ad+–W cannot be resolved due to the small

rotational constants (Ae=0.056, Be=0.028, Ce=0.028 cm@1

for Ia).

According to the calculations,the equilibrium structures

Ia and Ib have atoo high barrier for Wrotation around its b

axis (Vb&27 kJmol@1or 2250 cm@1,Figure S4) to explain the

observed (nearly) free rotation because of steric hindrance.

Hence,the small computed barriers between the three Ia and

Ib minima at TS1 and also TS2 suggest that in the effective

vibrationally averaged structure,the angle between the baxis

of Wand the C3axis of Ad+is much smaller than computed

for Ia/b.For example,the tilt angle for TS2 is 1.58

8leading to

anegligible rotation barrier. In this scenario,the Wligand

experiences only alow effective anisotropy for rotation

around its baxis because the three CH2groups at the top of

Ad+are far away,yielding low V3and V6parameters for an

effective one-dimensional internal rotation potential with

threefold symmetry.Calculations at other DFT levels confirm

that that the nonrigidity of Wabove the Ad+cation is the

reason for observing nearly free internal rotation (and not

awrongly computed equilibrium structure for Ia/b). We note

that free internal Wrotation has previously been observed for

anumber of cation–W clusters (A+–W), including small

inorganic (e.g.,A+=NH4+)[14] and atomic metal cations (e.g.,

Cr+/2+).[15] So far, free internal Wrotation has not been

identified for clusters with hydrocarbon cations,although the

IR spectra of the n3bands of A+–W with A=pentane[8b] and

protonated benzonitrile[13b] with linear CH···O and NH···O

ionic H-bonds show related spectral features (which were

however not analysed). In A+–W clusters with aromatic

hydrocarbons (e.g.,A=benzene or naphthalene), internal W

rotation is locked by bifurcated CH···O bonding to both lone

pairs of W.[10h,i] Thereason why the internal rotation in

Ad+–W is (nearly) free is because Wexperiences essentially

an almost spherical Ad+cation connected by asingle nearly

linear CH···O ionic H-bond.

TheC

@Hbond of Ad along the C3axis becomes strongly

elongated upon ionization (by 30 mc)and gets further

activated by monohydration (by 51 mc)upon formation of

the strong CH···O H-bond. Although the proton still remains

with Ad+,the large destabilization of the C@Hbond is

indicated by the large redshift of nCHtdown to around

2040 cm@1.Thus,asingle Wligand is not sufficient for driving

the intracluster proton transfer, which is necessary for the

functionalization reaction of Ad in apolar solvent. This

experimental result is consistent with our and previous DFT

calculations.[2d,3a] It is also in line with the thermochemical

expectation from the proton affinities of C10H15 (computed as

PA =868 kJmol@1), which is much higher than that of W

(691 kJmol@1).[12] From the proton affinities of W2and W3

(808 an 862 kJmol@1),[16] we expect proton transfer to solvent

to occur in (Ad–Wn)+clusters with n=2–3. Such astrong

activation of the aliphatic CH groups upon monohydration of

this prototypical cycloalkane cation has previously been

noted for linear alkane cations such as pentane (C5H12).[8b]

In (C5H12–W)+,the proton in the C···H+···O bond is already

closer to Othan to C, indicating that the C@Hbond in linear

C5H12+is even more acidic than in cyclic Ad+.Interestingly,

the more fundamental (CH4–W)+cation has the proton-

transferred form CH3–H+Wwith an OH···C ionic H-bond

(ROH···C =1.726 c,D0=69.8 kJmol@1,Figure S10), because

the PA of CH3is much lower than that of W(542 vs.

695 kJmol@1)leading to complete and barrierless intracluster

proton transfer from CH4+to W. This result is consistent with

the corresponding exothermic ion–molecule reaction of CH4+

with W(or CH4with W+)resulting in CH3and H+W.[17]

Finally,wecompare the strength of the CH···O ionic H-

bond in Ad+–W to those in related hydrocarbon radical

cations studied recently using the same computational and

spectroscopic approach. Thebifurcated CH···O H-bond

observed in polycyclicaromatic hydrocarbons (PAH) are

comparable to those in Ad+–W (D0=35.4 kJmol@1,RCH···O =

1.17 c,B3LYP-D3/aug-cc-pVTZ), such as benzene+–W

(38.4 kJmol@1,2.40 c)and naphthalene+–W (33.2 kJmol@1,

2.372 c), although their H-bonds are much longer and

involve less charge transfer to W(10 and 8mevs.

112 me).[10i] Consequently,they exhibit amuch smaller C@H

ngewandte

Chemie

Research Articles

12102 www.angewandte.org T2020 The Authors. Published by Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim Angew.Chem. Int. Ed. 2020,59,12098 –12104

bond elongation (2 and 1mcvs.42mc). Protonated

H+PAH–W cations have similarly strong bifurcated CH···O

H-bonds,asfor example in H+naphthalene–W (31.7 kJmol@1,

2.42 c)with small charge transfer (11 me) and minor C@H

bond activation (1 mc).[13a] On the other hand, the linear

C···H+···O H-bond in (C5H12–W)+is much stronger than in

Ad+–W,which is closer to aC

5H11–H+Wstructure,with an

enormous redshift of nCH down to 1300 cm@1and larger

redshifts in the free OH stretch modes.[8b]

Conclusion

In summary,the intermolecular interaction between W

and the Jahn–Teller distorted Ad+radical cation is charac-

terized by IRPD spectroscopy and DFT calculations of

Ad+–W to investigate the C@Hbond activation of this

prototypical diamondoid cation upon monohydration. Sig-

nificantly,these spectra provide the first spectroscopic

information of any microhydrated diamondoid cation cluster,

and thus result in afirst impression of activation of their C@H

bonds upon solvation with apolar solvent at the molecular

level. Thesalient results may be summarized as follows.The

IRPD spectrum of Ad+–W in the CH and OH stretch range

provides aclear picture of both the site and strength of

monohydration. TheWligand binds with one of its lone pairs

in astrong, short, and nearly linear CH···O ionic H-bond

(1.70 c)tothe single acidic CH group of the Jahn–Teller

distorted Ad+cation (isomers Ia/b), with acalculated binding

energy of D0=45 kJmol@1.Other isomers with other W

binding sites at the side and bottom of the Ad+cage (II,III)

and multiple CH···O contacts to three CH2groups are only

slightly less stable (DE0,5kJmol@1)but not observed

experimentally.The potential of Ad+–W is rather flat near

the global minimum leading to ahighly fluxional bonding

with little angular anisotropy.For example,the triply degen-

erate and nearly isoenergetic Ia/bminima with an energy

difference of only E0=0.3 kJmol@1are separated by low

barriers (Vb,3kJmol@1), and also structures with almost

parallel C2and C3axes of Wand Ad+are very low in energy

(<1kJmol@1). As aresult, the Wcan freely rotate around its

C2axis in an effectively quasi-planar CH···OH2configuration,

as documented by the rotational fine structure of the n3band.

Wattachment leads to further elongation of the acidic C@H

bond of Ad+already activated by ionization, causing alarge

redshift in the corresponding nCH frequency. However,

monohydration is not sufficient to drive intracluster proton

transfer from Ad+to W, in line with the proton affinities of

C10H15 and W. From the thermochemical point of view,two to

three Wmolecules are required for this process,which is the

basis for functionalization of Ad and other diamondoids in

polar solvents via aradical cation mechanism. To this end, the

analysis of IRPD spectra of larger (Ad–Wn)+clusters is

currently under way.Ingeneral, the CH···O bond in Ad+–W is

comparable to that in monohydrated PAH+and H+PAH

cations but weaker than in monohydrates of alkane cations,

such as pentane+.

Acknowledgements

This study was supported by Deutsche Forschungsgemein-

schaft (DO 729/8).

Conflict of interest

Theauthors declare no conflict of interest.

Keywords: adamantane cation ·C

@Hactivation ·hydration ·

IR spectroscopy ·structure elucidation

[1] a) P. v. R. Schleyer, J. Am. Chem. Soc. 1957,79,3292–3292;

b) R. C. Fort, P. v. R. Schleyer, Chem. Rev. 1964,64,277;c)J.E.

Dahl, S. G. Liu, R. M. K. Carlson, Science 2003,299,96–99.

[2] a) H. Schwertfeger, A. A. Fokin, P. R. Schreiner, Angew.Chem.

Int. Ed. 2008,47,1022–1036; Angew.Chem. 2008,120,1038–

1053;b)P.R.Schreiner,L.V.Chernish, P. A. Gunchenko,E.Y.

Tikhonchuk, H. Hausmann, M. Serafin, S. Schlecht, J. E. P. Dahl,

R. M. K. Carlson, A. A. Fokin, Nature 2011,477,308;c)W.L.

Yang,J.D.Fabbri, T. M. Willey,J.R.I.Lee,J.E.Dahl, R. M. K.

Carlson, P. R. Schreiner,A.A.Fokin, B. A. Tkachenko,N.A.

Fokina, W. Meevasana,N.Mannella, K. Tanaka, X. J. Zhou, T.

van Buuren, M. A. Kelly,Z.Hussain, N. A. Melosh, Z. X. Shen,

Science 2007,316,1460–1462;d)A.A.Fokin, P. R. Schreiner,

Chem. Rev. 2002,102,1551–1593;e)L.Wanka, K. Iqbal, P. R.

Schreiner, Chem. Rev. 2013,113,3516–3604;f)D.F.Blake,F.

Freund, K. F. M. Krishnan, C. J. Echer,R.Shipp,T.E.Bunch,

A. G. Tielens,R.J.Lipari, C. J. D. Hetherington, S. Chang,

Nature 1988,332,611 –613;g)R.S.Lewis,E.Anders,B.T.

Draine, Nature 1989,339,117–121;h)O.Pirali, M. Vervloet,

J. E. Dahl, R. M. K. Carlson, A. Tielens,J.Oomens, Astrophys.J.

2007,661,919–925.

[3] a) A. A. Fokin, P. R. Schreiner, P. A. Gunchenko,S.A.Pele-

shanko,T.E.Shubina, S. D. Isaev,P.V.Tarasenko, N. I. Kulik,

H. M. Schiebel, A. G. Yurchenko, J. Am. Chem. Soc. 2000,122,

7317–7326;b)M.Mella, M. Freccero,T.Soldi, E. Fasani, A.

Albini, J. Org.Chem. 1996,61,1413–1422.

[4] a) J. W. Raymonda, J. Chem. Phys. 1972,56,3912;b)S.D.

Worley,G.D.Mateescu, C. W. McFarland, R. C. Fort, C. F.

Sheley, J. Am. Chem. Soc. 1973,95,7580–7586;c)S.X.Tian, N.

Kishimoto, K. Ohno, J. Phys.Chem. A 2002,106,6541 –6553;

d) T. Rander, M. Staiger,R.Richter,T.Zimmermann, L. Landt,

D. Wolter,J.E.Dahl, R. M. K. Carlson, B. A. Tkachenko, N. A.

Fokina, P. R. Schreiner, T. Mçller, C. Bostedt, J. Chem. Phys.

2013,138,024310;e)A.Candian, J. Bouwman, P. Hemberger,A.

Bodi, A. G. G. M. Tielens, Phys.Chem. Chem. Phys. 2018,20,

5399–5406.

[5] A. Patzer,M.Schgtz, T. Mçller,O.Dopfer, Angew.Chem. Int.

Ed. 2012,51,4925–4929; Angew.Chem. 2012,124,5009–5013.

[6] a) A. Gali, T. Demj#n, M. Vçrçs,G.Thiering,E.Cannuccia, A.

Marini, Nat. Commun. 2016,7,11327;b)T.Xiong,R.Włodarc-

zyk, L. Gallandi, T. Kçrzdçrfer, P. Saalfrank, J. Chem. Phys.

2018,148,044310;c)T.Xiong,P.Saalfrank, J. Phys.Chem. A

2019,123,8871–8880.

[7] O. Pirali, V. Boudon, J. Oomens,M.Vervloet, J. Chem. Phys.

2012,136,024310.

[8] a) M. Xie,Y.Matsuda, A. Fujii, J. Phys.Chem. A 2016,120,

6351–6356;b)T.Endo,Y.Matsuda,A.Fujii, J. Phys.Chem.

Lett. 2017,8,4716–4719.

[9] a) O. Dopfer, Int. Rev.Phys.Chem. 2003,22,437–495;b)O.

Dopfer, Z. Phys.Chem. 2005,219,125–168;c)E.J.Bieske,O.

Dopfer, Chem. Rev. 2000,100,3963–3998.

ngewandte

Chemie

Research Articles

12103Angew.Chem. Int.Ed. 2020,59,12098 –12104 T2020 The Authors. Published by Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim www.angewandte.org

[10] a) O. Dopfer,D.Roth, J. P. Maier, J. Am. Chem. Soc. 2002,124,

494–502;b)N.Solc/,O.Dopfer, Angew.Chem. Int. Ed. 2002,

41,3628–3631; Angew.Chem. 2002,114,3781–3784;c)N.

Solc/,O.Dopfer, Angew.Chem. Int. Ed. 2003,42,1537 –1540;

Angew.Chem. 2003,115,1575–1579;d)A.Patzer,S.Chakra-

borty,N.Solc/,O.Dopfer, Angew.Chem. Int. Ed. 2010,49,

10145–10148; Angew.Chem. 2010,122,10343–10346;e)H.S.

Andrei,S.A.Nizkorodov,O.Dopfer, Angew.Chem. Int. Ed.

2007,46,4754–4756; Angew.Chem. 2007,119,4838–4840;

f) H. S. Andrei,N.Solc/,O.Dopfer, Angew.Chem. Int. Ed. 2008,

47,395–397; Angew.Chem. 2008,120,401–403;g)K.Chatter-

jee,Y.Matsumoto,O.Dopfer, Angew.Chem. Int. Ed. 2019,58,

3351–3355; Angew.Chem. 2019,131,3389–3393;h)K.Chat-

terjee,O.Dopfer, Chem. Sci. 2018,9,2301–2318;i)K.Chatter-

jee,O.Dopfer, Phys.Chem. Chem. Phys. 2017,19,32262–32271.

[11] M. J. Frisch, et al.,GAUSSIAN09, Rev.D.01, Gaussian, Inc.,

Wallingford CT, 2009.

[12] P. J. Linstrom, W. G. Mallard, NIST Chemistry WebBook, NIST

Standardsand Technology,Gaithersburg MD,20899 (http://

webbook.nist.gov), 2001.

[13] a) K. Chatterjee, O. Dopfer, J. Phys.Chem. A 2020,124,1134–

1151;b)K.Chatterjee, O. Dopfer, Phys.Chem. Chem. Phys.

2019,21,25226–25246.

[14] a) P. J. Kelleher,C.J.Johnson, J. A. Fournier,M.A.Johnson,

A. B. McCoy, J. Phys.Chem. A 2015,119,4170–4176;b)T.

Pankewitz, A. Lagutschenkov,G.Niedner-Schatteburg, S. S.

Xantheas,Y.T.Lee, J. Chem. Phys. 2007,126,074307.

[15] P. D. Carnegie, B. Bandyopadhyay,M.A.Duncan, J. Phys.

Chem. A 2008,112,6237–6243.

[16] a) D. J. Goebbert, P. G. Wenthold, Eur.J.Mass Spectrom. 2004,

10,837–845;b)R.Knochenmuss,O.Cheshnovsky,S.Leutwyler,

Chem. Phys.Lett. 1988,144,317 –323.

[17] a) F. W. Lampe,F.H.Field, J. L. Franklin, J. Am. Chem. Soc.

1957,79,6132–6135;b)D.Smith, N. G. Adams, Int. J. Mass

Spectrom. Ion Phys. 1977,23,123–135;c)W.T.Huntress,R.F.

Pinizzotto, J. B. Laudenslager, J. Am. Chem. Soc. 1973,95,4107–

4115;d)I.Dotan, W. Lindinger, B. Rowe,D.W.Fahey,F.C.

Fehsenfeld,D.L.Albritton, Chem. Phys.Lett. 1980,72,67–70.

Manuscript received:March 11, 2020

Acceptedmanuscript online: May 11, 2020

Version of record online: June 4, 2020

ngewandte

Chemie

Research Articles

12104 www.angewandte.org T2020 The Authors. Published by Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim Angew.Chem. Int. Ed. 2020,59,12098 –12104