1  

Marek  Pruski  

 U.S.  DOE  Ames  Laboratory,  Ames,  Iowa  50011,  U.S.A    

 Department  of  Chemistry,  Iowa  State  University,  Ames,  Iowa  50011,  U.S.A.  

Emerging  trends  in  NMR  of  materials  

UHF Workshop, Bethesda, November 12, 2015

2  

Research  needs  in  materials  science  

Understanding  the  structure-­‐property  rela:ons  in  novel  materials  is  an  unanswered  scien:fic  challenge,  which  prevents  ra:onal  design.  A  few  examples:  !   heterogeneous  catalysts:  supports,  cataly8c  sites,  reac8on  

mechanisms  !   energy  related  materials:  ba<eries,  fuel  cells,  hydrogen  storage,  

thermoelectrics,  photovoltaics,  …  !   construc8on  materials:  cements,  polymers,  composites  !   glasses  and  ceramics  !   semiconductors,  ion-­‐conduc8ng  solids,  and  dielectric  materials  !   materials  for  environmental  remedia8on  and  CO2  sequestra8on  !   biomaterials  !   pharmaceu8cals  (more  examples  are  given  in  Task  Force  Summaries,  p.  25-­‐26)    

3  

Solid-­‐state  NMR  

Challenges:    !   intrinsically  low  sensi8vity  of  NMR  !   low  resolu8on;  homogeneous  and  inhomogeneous  line  

broadening  in  solids  

SS-­‐NMR  spectroscopy  has  an  unparalleled  ability  to  provide  atomic-­‐level  characteriza:on  of  materials:    !   most  elements  have  NMR-­‐ac8ve  isotopes    !   nuclear  spins  are  excellent,  site-­‐dependent  “reporters”  of  local  

structure  and  dynamic  processes    

Transforma:onal  role  of  ultrahigh  field  in  SS-­‐NMR:  !   ultrafast  MAS;  indirect  detec8on  !   low-­‐gamma  nuclei    !   half-­‐integer  quadrupolar  nuclei  !   high-­‐field  DNP  

4  

Characteriza:on  by  SS-­‐NMR  

1D MAS spectra of catalytic surface; 14.1 T, MAS at 40 kHz

-150 -100 -50

1H

H1

10.0 7.5 5.0 0.0 2.5

H2

H3 C3

0 20 40 60 80 100 120 140 160

C2 C1

CTAB MeO

13C 29Si OH/H2O

T3 Q2

Q3

Q4 T2

-60 -80 -100 -120 0

2 4

6 8

-1 0 1 2 3 4 5 6 7 8

-1 0 1 2 3 4 5 6 7 8

2D correlation spectra of the same system

1H-13C 1H-1H 1H-29Si

T3 Q3

Q4

T2

H1

H3

H2

H1

H3

H2

C3

0

1

2

3

4

5

6

7

dH ppm

0 20 40 60 80 100 120 140 160 dC ppm

C2 C1

H1

H3

H2

CTAB MeO

AL-MSN 1 3

2

Si

J. Trebosc, et al., J. Am. Chem. Soc., 2005, 127, 3057-3068; J. Am. Chem. Soc., 2005, 127, 7587-7593.

5  

Emerging  technique:  ultrafast  MAS  

Transforma:onal  role  of  ultrafast  MAS  in  SS-­‐NMR:  

New probes (e.g. ultrafast MAS): new pulse sequences/theory; improved resolution & sensitivity

!    MAS  rate:    100+  kHz            !    volume:      ~0.3  μl    !   introduced: ~2012

!    MAS  rate:    45  kHz    !    volume:      ~9  μl    !   introduced: ~2005

!   improved  1H  resolu8on    (Δν~(νMAS)-1)  

!   indirect  detec8on  !   sideband-­‐free  spectra  

!   be<er  sensi8vity/spin  !   higher  RF  fields  !   improved  decoupling/recoupling  !   dipolar  trunca8on  

Need higher magnetic field!!!

6  

1H  resolu:on  under  ultrafast  MAS  

L-histidine HCl H2O

10.0 010.0 010.0 010.0 010.0 010.0 0

Y.  Nishiyama,  JEOL  Resonance.    

1H MAS at 14.1 T

80 kHz 100 kHz

60 kHz

40 kHz

20 kHz

110 kHz

!   1H  resolu8on  and  SNR  are  greatly  enhanced  by  fast  MAS  !   CRAMPS-­‐like  resolu8on  approached  at  100  kHz  

7  

1H  resolu:on  under  ultrafast  MAS  

L-histidine HCl H2O

Y.  Nishiyama,  JEOL  Resonance.    

1H INADEQUATE at 14.1 T

100 kHz

110 kHz

90 kHz

80 kHz

60 kHz

120 kHz

!   1H  resolu8on  and  SNR  are  greatly  enhanced  by  fast  MAS  

8  

1H-­‐1H  correla:ons  under  MAS  at  100  kHz  

100 kHz

80 kHz

60 kHz

40 kHz

-25

25

20

-20

-15

-10

-5

0

5

10

15

δ H(D

Q, p

pm)

10 5 0 -5 -10δH (SQ, ppm)

10

-10

-5

0

5

δ H(ppm)

10 5 0 -5 -10δH (ppm)

10

-10

-5

0

5

δ H(ppm)

10 5 0 -5 -10δH (ppm)

1D MAS

1H-1H DQMAS 100 kHz

slice

A

B

B

A

A-B

B

A 1H-1H spin diffusion 60 kHz

T.  Kobayashi,  et  al.,  Angew.  Chem.  Int.  Ed.,  2013,  52,  14108.    S.P.  Brown,  Macromol.  Rapid.  Commun.  2009,  30,  688-­‐716  

Ultrafast  MAS  at  100  kHz:  improved  resolu:on  in  2D  1H-­‐1H  NMR  !  Host-­‐guest  interac8on  in  corrole/toluene  system  

Polymers, supramolecular systems, catalysts, …

9  

Indirect  detec:on  of  low-­‐γ  nuclei

1H

Low-γ

S/N gain: ( )( )

2/32/1

//

⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛≈

X

H

H

X

DD

ID

NSNS

γγ

δνδνα

Tradi:onal  SSNMR  approach:  direct  detec:on  of  low-­‐γ nuclei  (e.g.  13C,  15N)  

1H

Low-γ

Indirect  detec:on  of  low-γ nuclei    

!   1H homonuclear RF decoupling !   low sensitivity

!   1H decoupling by fast MAS !   high sensitivity

For 15N:

�

γ HγN

⎛

⎝ ⎜

⎞

⎠ ⎟ 3 / 2

= 31 time performance improves by ~103!

t2 : 1H

t1 : 1H

t2 : X

t1 : X

10  

1H-­‐13C  HETCOR  under  fast  MAS:  cataly:c  surface  

8 6 4 2 0

0 50

10

0 15

0

150 125 100 75 50 25 0

0 2.

5 5

7.5

H1 H3 H2

C2

C1

C3

CTAB

CTAB

(b) C2 C3 C1

H1

H3

H2

CTAB

CTAB (a)

Si≡ 1

2

3

F1

F2

F1

F2

Direct 15 h

Indirect 15 min

AL-MSN 1 3

2

Si

13C, 15N

CP or INEPT

t1

Y Y,Y

t2 SPINAL-64

1H

X

Y

Decoupling

2τRR

ϕ, ϕ+π

τCP τCP

X Y

Recoupling

X

!  1H-1H homonuclear RF decoupling INEPT

Through space correlations: Ishii, Y.; Tycko, R. JMR, 2000, 142, 199; Wiench, J.W. et al. JACS 2007, 129, 12077; Zhou D.G. et al., JACS 2007, 129, 11791

Through-bond correlations: Elena, B. et al. JACS, 2005, 127, 17296; Mao, K.; Pruski, M. JMR, 2009, 201, 165;

40 kHz MAS

11  

1H-­‐13C  HETCOR  under  MAS  at  100  kHz    

!   Excellent  sensi8vity  and  resolu8on;  all  resonances  detected        

!   1H-­‐1H  interac8ons  are  suppressed  during  CP;  one-­‐bond  selec8vity  (dipolar  trunca8on)  

!   HMQC  possible  when  T2’  >  10  ms  

5 4 3 2 1

40

30

20

10

13C

Che

mic

al S

hift

/ ppm

1H Chemical Shift / ppm

10 8 6 4 2 0 -2 -4

140

120

100

80

60

40

20

0

13C

Che

mic

al S

hift

/ ppm

1H Chemical Shift / ppm

MP

a

b

c

a

b c

7 h

MP-­‐MSN  

T.  Kobayashi,  et  al.,  Angew.  Chem.  Int.  Ed.,  2013,  52,  14108.    

12  

!   Sensi8vity/scan  at  100  kHz  close  to  that  of  1.6-­‐mm  probe  (>50%  in  terms  of  SNR),  in  spite  of  20x  smaller  volume  

!   Resolu8on  in  (A)  similar  to  PMLG  (C)    

HSQC  of  f-­‐MLF-­‐OH  under  natural  abundance  at  14.1  T  

Y.  Nishiyama,  T.  Kobayashi,  et  al.,  SSNMR,  66-­‐67,  56-­‐61  (2015)    

(A)  100  kHz  MAS  (0.75-­‐mm);  1H  detec:on;  5  h    

(B)  41.667  kHz  MAS  (1.6-­‐mm);  1H  detec:on;  10  h    

(C)  41.667  kHz  MAS  (1.6-­‐mm);  13C  detec:on;  PMLG  during  t1,  10  h    

13  

13C  CPMAS  NMR  of  Argonne  Premium  Coals  at  low  and  high  magne:c  field  

 

8 mg 8 h 42 kHz 1 +

Sample weight: Acquisition time: MAS rate: Sensitivity (per scan): Resolution:

150 mg 8 h 8 kHz 1.5-2.0 - 15N or 33S??

14  

Studies  of  low-­‐γ  nuclei:  15N  under  natural  abundance  

Similar  techniques  can  be  used/developed  for  other  low-γ nuclei  in  many  classes  of  materials  

15N-­‐1H  HETCOR  of  amino  acids        through  space                through  bond

Need higher magnetic field!!!

Althaus  et  al.,  SSNMR,  57-­‐58,  17-­‐21  (2014);  collabora8on  with  R.  Schurko,  University  of  Windsor    

15N-­‐1H  HETCOR  of  pharmaceu8cals  through  space            

�

γ HγN

⎛

⎝ ⎜

⎞

⎠ ⎟ 3 / 2

= 31

15  

Half-­‐integer  quadrupolar  nuclei  Most NMR-active nuclei are half-integer quadrupolar: 77 out of ~110, including 7Li, 11B, 17O, 23Na, 25Mg, 27Al, 33S, 35Cl, 39K, 43Ca, 55Mn, 87Rb, … The challenge: SSNMR spectra are dominated by quadrupolar broadening

ΔE(2)  

ΔE(1)  

ωcentral

ΔE(1)  

m  

+3/2  

+1/2  

-­‐1/2  

-­‐3/2  

     Zeeman I-st order II-nd order          

3ω0  

ω0  

ω0  

ω0  

ω0  

)1cos9)(cos31}(4/3)1({16

22

0

2

0 −−−+−= ββω

ωωω IIQ

central

Typical values for 27Al: = 1 MHz, = 10 kHz 27Al chemical shift range at 9.4 T: 10 kHz

)1(mEΔ )2(

mEΔ

ω0  

I-­‐st  ordrer  single    crystal  

II-­‐nd  order  (CT)  

Sta8c  spectra  

I-­‐st  ordrer  powder  

16  

Advances  in  NMR  of  quadrupolar  nuclei:  complete  line-­‐narrowing  

isotropic  

anisotropic  

ω−m↔m(2) =

ωQ2

ω 0

A0 (I , p)B0Q (ηQ )+

ωQ2

ω 0

A2 (I , p)B2Q (ηQ ,αQ ,βQ )P2 (cosθ )+

ωQ2

ω 0

A4 (I , p)B4Q (ηQ ,αQ ,βQ )P4 (cosθ )

csmmmmmm m ↔−↔−↔− ++= ωωωω )2(

02

)](cos),,([2 220 θβαηδωω PBm cscscscscs

mm +Δ=↔−

where:  p = 2m; αQ, βQ –  angles  between  QPAS  and  rot.  axis;  θ    -­‐  angle  between  rota8on  axis  and  B0;  P2,4(cosθ) –  2nd  and  4th  order  Legéndre  pol.;  At(I,p), i = 0,2,4 -­‐  coefficients  The  resonance  frequency  also  includes  chemical  shit  

P2(cosθ) and  P4(cosθ) do  not  have  a  common  root;  MAS  narrows  second  order  broadening  only  by  a  factor  of  ~3;  DOR,  DAS  and  MQMAS  yield  isotropic  spectra  

with    

A. Pines et al., Mol. Phys. 65, 1013 (1988); JMR, 86, 470 (1990); L. Frydman et al., JACS, 117, 5367 (1995)

}Symmetric  transi8on  under  fast  rota8on  of  the  sample  

17  

Example:  mechanisms  of  dehydrogena:on  in  metal  hydrides  

Mechanochemistry  

DFT  Modeling  Solid-­‐state  NMR  7Li,  11B,  23Na,  27Al  

18  

Example:  mechanisms  of  dehydrogena:on  in  metal  hydrides  

Need higher magnetic field!!!

Our  approach:    !   measure  1D  and  2D  NMR  spectra  of  1H,  

7Li,  11B,  23Na,  27Al  and  other  nuclei  in  hydrides  processed  under  various  condi8ons  and  in  reference  compounds  

!   obtain  chemical  shits  and  quadrupolar  parameters  (for  spins  >  1/2)  to  iden8fy  the  coordina8on  geometries  and  chemical  structures  

!   carry  the  out  addi8onal  SSNMR  experiments  to  probe  interatomic  correla8ons,  molecular  mo8ons,  etc.  

!   refine  the  structures  using  molecular  modeling  and  DFT  calcula8ons  

11B  MQMAS  

11B  MAS  

DFT  (δCS,  PQ)  

19  

Spectra  of  quadrupolar  nuclei  at  ultrahigh  fields  

Transforma8onal  role  of  ultrahigh  field:  at  40  T,  second  order  quadrupolar  broadening  and  shit  are  diminished;    the  line  width  is  mainly  due  to  field  drit  

27Al  MAS  spectra  of  aluminoborate  9Al2O3+2B2O3              

Simulated  MAS  spectra  of  23Na  in  NaC2O4/NaSO4  mixture  at  14.1  T  and  36  T    

40  T  (NHMFL,  hybrid)  

25  T  (NHMFL,  resis:ve)  

19.6  T  

Z. Gan et al., JACS, 124, 5634 (2002)

ΔEm(2),ω−m↔m

(2) ∝ B0−1

14.1  T  

20  

2D  27Al-­‐31P  HETCOR  spectra  of  AlPO4-­‐14:    

!   Resolu8on  of  MAS-­‐based  spectra  at  18.8  T  in  (c)  and  (d)  rivals  that  of  MQ-­‐HETCOR  taken  at  9.4  T!    (not  shown  here)  Assuming  20%  efficiency  of  MQMAS,  8me  performance  at  36  T  could  improve  by  an  incredible  factor  of  (36/9.4)5  8mes  52  ≅  2x104    

2D  HETCOR  spectra  of  quadrupolar  nuclei  at  high  fields    

!

M. Pruski et al., JMR, 184, 1 (2007)

9.4T

18.8T 18.8T

18.8T 18.8T

9.4T

21  

Conclusion  and  acknowledgments  

Funding  (Ames)  U.S. Department of Energy, BES (DE-AC02-07CH11358)

Coworkers/collaborators        Ames:    

T.  Kobayashi    I.I.  Slowing    J.W.  Wiench    V.K.  Pecharsky  V.  Lin      D.  Johnson  S.M.  Althaus      S.  Gupta  

   

Tokyo:    Y.  Nishiyama  (JEOL)  

 Lille/Caen:      J.-­‐P.  Amoureux  

   C.  Fernandez    J.  Trebosc  

Thank  you  DOE,  NSF  and  NIH  for  suppor:ng  our  science!    

The  role  of  ultrahigh  field  in  SS-­‐NMR  and  materials  research  will  be  transforma:onal:  

!   drama8cally  expanded  capabili8es:  resolu8on,  detec8on  limits,  range  of  nuclei  

!   drama8cally  expanded  range  of  applica8ons  to  new  materials    

ExxonMobil:    K.  Mao  G.  J.  Kennedy