Transcript Spectroscopy in Organic Chemistry….

Welcome to Chemistry 421
• Use of Physical Methods to determine
structures in Organic Chemistry
Dr. Charles DeBrosse
232 Beury Hall (across from Dept. office)
NMR lab 001 Beury Hall
215-204-1082
Course Organization
• Lectures Tuesday evenings 6-9pm. Attendance is
strongly encouraged. You should read the assigned
material and do the homework problems prior to
the lecture.
• Textbook, Organic Structural Spectroscopy, by
Lambert, Shurvell, Lightner and Cooks
• Grade based on 3 In-class quizzes, homework sets
and Final exam
• Course notes will be available on Blackboard
generally the Monday before the lecture.
• I will reserve Monday 10-11am as office hour, and
am accessible other times in my lab or office
Spectroscopy in Organic
Chemistry….
•
•
•
•
•
•
The Chemists Eyes, Ears and Nose
How do we know what we have?
Read labels (sometimes labels lie)
Whose word do we have to take on it?
Check it out for yourself! (Get a spectrum!)
A spectroscopist. (never the hero, just the
hero’s best friend)
Chemistry 421
• Goal of the course is to give you the tools to
answer parts of these questions.
• Organic Structure determination by spectroscopic
methods
• NMR (nuclear magnetic resonance spectroscopy)
• Mass Spectrometry (MS)
• Infrared Spectroscopy (IR and other vibrational
classes like Near IR, Raman)
• Electronic Spectra (Ultraviolet and chiroptical
methods)
Structural Features we can
address Spectroscopically
•
•
•
•
•
Molecular weight
Chemical Formula
Functional groups
Skeletal Connectivity, structural isomers
Spatial-geometric arrangements, stereoisomerism,
symmetry
• Presence and location of chromophores
• Chirality issues
• Some of these are more central than others.
Sometimes we can stop when the answer is fit to
purpose.
Techniques we will study in this
Course
• NMR--looks atb atoms by means of their nuclei.
Connectivity pathways, spatial arrangements of atoms and
1:1 correspondence between signals and atoms
• Mass Spec--measures molecular weight, most
fundamentally useful for unknowns. Controlable
fragmentation can distinguish among rival possibilities
• IR and Raman--Vibrations characteristic of bonds,
particulary for functional group identification. Excellent
“fingerprint”
• UV--reports on conjugation and multiple bonds. Provides
entre’ to chiroptical probes to assymetric configurations
Chemistry 421
• Goal is for you to gain a conversational
level of knowledge
• Base level of theory, as pictoral as possible
should help you realize the scope and
limitations of these methods.
• Survey of representative data and how to
interpret it.
• Applying the right tool to the right question
Integrating data from various tools and sources (i.e. we
should be able to explain all our observations. Also we
should be able to observe features we predict, knowing
our chemistry)
Complementary Spectroscopies
•There are strengths and weaknesses in all the various spectroscopy methods
•“Blind Spots” (see story of three blind men describing an elephant)
•NMR e.g.; has “NMR silent” nuclei; fully substituted carbons as “blocks”; blind
to inorganics, nearly blind to polymeric mtls.
•IR e.g.; robust for functional groups; great if you have a compound match in a
library (fingerprint); shaky on quantitative response, same spectrum might
explain multiple compounds; subject to selection rules governed by dipole
moment in vibration (complement somewhat with Raman)
•UV; needs chromophores
•Mass Spec; compounds differ in ionizability.
A smart chemist will be attuned to possible blind spots before making
conclusions. Best solution is to “marry up” complementary data. All
the data needs to “agree” or at least, not conflict withing the various
methods.
Total Synthesis as a Structural
Proof
• Use of all our knowledge
• Generally the product of a reaction is
rationally related to the ingredients we have
used.
• Non-ambiguous route from known
compounds
• Oxidative degradation to known compounds
and history of chemistry
The Electromagnetic Spectrum
•Light comes in different colors
•No matter what part of range, there are some features in
common, that you should know.
B
B
1 wavelength, 
is a distance….
A
 is frequency, Hz, 1/sec
Amplitude
PhotonE= h
 = c/; |c=3x1010 cm/sec
h = Planck’s constant = 6.624 x10-34 J•sec
/c = 1/  = wavenumber, cm-1
Peaks per time;
frequency
Propagation of e,b fields, time
Chemical Properties as related to
the different colors of light
Spacings
between atomic
nuclei in crystals
higher E,
higher 
smaller 
-15
lower E,
lower 
larger 
Micro
X- UV IR+Ramanwaves NMR
RayElectron
vibrations of
orbital
bonds
transitions
-10
-5
100
kcal/mol
10
kcal/mol
rotational nuclear
states
spin flips
0
5
10-6
kcal/mol
Log ,
meters
What about Mass Spec?
The properties measured do not directly bear on absorption of
light
More properly termed mass spectrometry
We do however in quadrupole mass spectrometry scan an
electric field that induced different curve paths for different
masses
Radiofrequency also does show up in FT mass spec (ion
cyclotron resonance)
Sensitivity of the methods we
will study
• NMR is worst, typically needing 10s of g
to 10s of mg for 13C.
• Mass Spec is generally about 103-104 x
more sensitive than NMR
• UV is about 100 x more sensitive than
NMR
Questions, Questions….
•
•
•
•
•
•
•
•
•
What’s in this flask?
What have you just synthesized?
Did my reaction Work?
What is present in that sample?
How could this reaction have possibly Failed?
Isolated materials?
Purity? Mixture? What kind of mixture?
Is this material suitable for the next step?
Best to Ask Yourself these questions! (Better than
making your Boss ask, or not to be able to answer
a “Customer”
Answers!
• Bad Answer: “…because this always works in our
lab…”
• Bad Answer #2 “..because my Professor (or some
slightly older grad student, postdoc) said so..”
• Slightly less Bad Answer #3 “…this was done in
the literature…”
• Good Answer “…because all the spectral and
other analytical data agree with me”
• Analytical Chemistry. If you do it right, nobody
has to “take your word” on the answers!
Features in Common for all
Spectroscopy
Some Physical
Property
Energy
Transitions
Light,
(Frequency
Dispersion)
A key for us here is, we use
instruments to Disperse energy
across a scale appropriate to a
chemical property
Measuring
Scheme
Data
Analysis,
Interpretation
Knowledge
Wisdom (or Progress)
Therefore we have Precise
Analytical Instruments that can
Disperse Energy
•Accurate, precise, reproducible
•Combine the energy dispersion scheme with a detection
scheme.
•Generally the sample sits physically between the source
and the detector.
•Detector provides selectivity in response, usually generates
a voltage. We record voltage responses as “DATA”
The dispersion is easy to achieve
with ordinary light
Light source
Monochromator rotates
Prism or diffraction
grating
Spacings on grating
appropriate to
wavelength
Recorder
Sample
Detector
Schemes use slits to
admit a select region of
spectrum
Pretty ineffective for
radio waves
Spectroscopy “Spreads out Vision”
All the techniques we will discuss have some features in
common
Data will have a running variable (x-axis) that is in some sense,
a “energy” scale. (not at least directly, a time axis. Therefore a
snapshot in time of a molecule)
Instrument
Response
The response variable (an absorption or other intensity) is
related to the chemical preponderance of some feature that cause
the response.
Energy
There is a peak here
But not here
The position informs us
about some chemical
property in the sample
The peak height
informs us about how
much of that property
is in the sample
Fourier Transform Methods
An alternative to Energy
Dispersive methods
•All modern NMR and IR is done this way
•Measures all frequencies at same time. More efficient at
signal-gathering in a give time (better S/N)
•The frequencies present are deconvoluted (or dispersed) after
data is collected.
•Fourier Analysis is the mathematical method for doing this. It
is based on the theory that any complex periodic (repeats over
time) wave can be decomposed into a linear combination of
sinusoids
Instrument
Response
To get the measurement, we
collect a detector response as a
function of time
time
Lots of different frequencies present from the sample
Their voltages “beat” against each other making interference pattern
(interferogram)
Interference is periodic, because the frequencies are constant w.r.t each other
An Oscillating voltage is interpreted as a Frequency
The process is similar
to the way a sound
wave is digitized to
make e.g. a music CD
Key to this is
sampling at
exactly equal time
intervals
This is a Frequency
Axis. Think Hz!
Interfering Sinusoids are
Represented in a decaying trace
Space is frequency 1
Space is beating of frequency 2 vs 1 (1 - 2)
A human being
could compute
this FT, counting
beats per time
unit
Key to the process is a very precisely
defined time base (the x axis) that the FT
algorithm uses to count
Interference patterns--Almost
able to Transform by Hand…
Damped Sines of 2.77, 28, 34 Hz
Added together
2.5
1
2
1.5
0.5
1
0.5
0
0
-0.5
-0.5
-1
-1.5
-1
0
20 0
40 0
60 0
80 0
10 00
0
200
400
600
800
Time(ms)
Time(ms)
But its really the Fast Fourier Transforms and fast computers that make this
practical!
1000
Since the time-acquisition is fast and
efficient it is easy to Signal-Average
Adding accumulating scans from the detector into memory of
computer
Signals are coherent and adding the scans causes signal to
grow linearly with number of scans.
Noise being random and incoherent grows with √no.of scans
From this, the Signal-to-Noise ratio (S/N) grows
proportionally to the square root of number of scans
E.g., a spectrum acquired with 100 scans will be 10x better
than one with 1 scan only.
Some Features common to all
Spectra
Linewidth. Usually measured at
0.5 Imax
Can be limited by the
instrumentation, or be limited by
nature.
Nature, here exerts herself as
uncertainty to to slight
chemical variation, or inability
to measure energy precisely.
Units of linewidth are same as
axis, e.g. Hz, cm-1
Resolution (ability to distinguish
line from closely spaced
neighboring line is related to
linewidth
Using computerized
Data systems adds an
additional limitation
on our resolution
problem
More data points are
better but usually at a
cost to expt efficiency
See here 2 identical
peaks, digitized
differently
Bears directly on our
ability to determine
the position of Imax
Noise, the curse of Science
•All measurements, especially those we carry out with
instruments, generate Noise.
•Detectors of all sorts generate electrical noise
•Noise is bad. It is random and incoherent and does not
possess information. We go to tremendous expense and effort
to eliminate, suppress, and finesse our way past noise.
•Signals are good. They give us information.
•Noise limits our ability to even observe very weak signals or
to quantify somewhat weak signals. The Signal-to-Noise
Ratio is an important parameter is assessing our ability to
interpret data.
•Noise is superimposed on top of peaks
Signal-to-Noise
(S/N) ratios
Typical rule of thumb:
Measure
height
Limit of detection, S/N=3
Limit of Quantitation, S/N=10
Noise(rms) is
S/N=6.3/2*0.707
0.707 x peak to
=4.45
peak
So this peak is reliably
detectable, but not
reliably quantitatable
Chemistry 421--Structure
Determination
• Interpretable Connection between Structural
Features and Spectroscopic signals
• We will interpret spectra to learn about
structures.
• The Interpretation “paradigm” consists of
charting:
Observation
Inference
IR shows a C=O at 1730cm-1
My molecule contains a ketone or
aldehyde
I may have a methyl group
adjacent to C=O
NMR shows a 3-proton singlet at
2.2
May need
more
experiments
or to look
further in the
data!
A way of Thinking……
????????
Known compounds
Total Unknowns
(verify structure)
Isolated natural products
All predicted signals present?
Unrelated impurities,
contaminants
Agreement with literature?
Impurities present?
Fingerprint?
“Partial Unknowns”
New compound?
Side reaction product
Wrong starting material, carry
through known synthetic steps
Can we track known compounds?
Peaks from precursor compounds
may have “descendents”
Single component, vs
mixture?
A Strategy for Handling
Unknown Structures
• Complementary 1H NMR, 13C NMR, Mass Spec, UV--any
features stand out?
• Get the Molecuar Weight from MS
• Heavy Atoms? (ratio of M to M+1, M+2)
• If heavy atoms are identified, subtract from MW
• Consult various molecular formula DBs (Merck, CRC etc). Write
out Molecular Formula
• Use the DBE (sites of unsaturation) rule
• Infrared-- Functional groups present? Identify as possibly subtract
from formula (retain the need to incorporate at end)
• Inventory 13C NMR and classify the C,H groups present. Tabulate
fragments of structure. Reconcile MS fragments.
• Assemble possible structures
Molecular Weight and Molecular
Formulas
•Absolutely critical to Stucture determination
•Centrality of Mass Spectrometry to modern Chemistry
•Molecular weight must agree with the structure. Note well,
that a given nominal MW generally is consistent with several
possible formulas.
•The “nitrogen rule”. A compound with an even-numbered
molecular weight has 0, 2 or an even number of nitrogens.
•Very Important: Learn the rule for sites of unsaturation
(double-bond equivalents, DBE) as a predictive tool for
multiple bonds and/or rings. These are based on the standard
valencies for ordinary atoms.
DBE Rules
• Aim to reduce a formula to CNH2N+2
• Take formula and cross off Oxygen atoms
• Replace all halogen atoms with hydrogen
• Cross off all Nitrogen atoms, and for each N
remove one H atom.
• Sulfur treat like Oxygen (? Use care if there are a lot
of oxygens, possible O=S=O type groups, similar
issues with Phosphorus)
• Subtract your newly reduced formula (looks like
CxHx, from CxH2x+2
number H (even number)
• Divide this answer by 2. Result is DBE.
So some “Decision-Tree”
thinking is possible The NMR
branch.
Data
“granularity” of questions
Synthetic Product
1H NMR
Could it be what I want?
Quick Inventory of signals
YES
NO
Do I need more information?
What do I need to find out
Carbon Survey
Need Assignments?
Proton coupling pattern
YES
Integrate at
higher level
with other
techniques
Worth more spectroscopy?
Back to the Lab!
Assess Purity
NO
Correlations to protons
Information
Content higher
Noe for stereochemistry
Separations
methods, Feedback
to synthesis.
Nuclear Magnetic Resonance
(NMR)
Tonight’s Subjects
I.
How do the spectrometers work?
II.
The NMR measurable quantities
What is NMR Spectroscopy?
• Nuclear Magnetic Resonance
• Radio Frequency Absorption Spectra of atomic nuclei in
substances subjected to magnetic fields.
• Spectral Dispersion is Sensitive to the chemical
environment via “coupling” to the electrons surrounding
the nuclei.
• Interactions can be interpreted in terms of structure,
bonding, reactivity
The Fundamental NMR
equations
• Spinning nuclei produce a magnetic field that is
proportional to its magnetic moment . The
proportionality constant is ;  = hI
• An active nucleus in a magnetic field B0 has an energy
w.r.t. zero field of:
E (= h= h) = - • B0 where  is the component of the
magnetic moment colinear with B0
This gives for IZ= ±1/2; E = ± 1/2 (h B0)
E = h B0 and in angular units  =  B0
Origin of the NMR Effect
• Nuclei with other than A(#protons+neutrons) and
Z(#protons) both even numbers, possess net spin and
associated angular momenta
• Reveals itself only in magnetic field. As usual, such
momenta are quantized
• States have different energies, populated according to
Boltzmann distribution
• States are 1/2, 3/2, 5/2…for A= odd number and integer if
A= even number and Z= odd number
• Transitions of individual nuclei between spin states is
possible (both directions) leading to an equilibrium of
populations
• Number of states is 2I + 1
Pictoral View of Spin
Direction of the Applied
Magnetic Field
Averages out in x,y plane; small
net resultant vector along the
z axis
Z

Nucleus Moments
precess about magnetic
field.
X
Quantized either with or
oppossed to field
Z
M
The Boltzmann excess of
low over high energy
state is very small, 1 in
106
Y
X
Y

Precession of nuclear
magnet--Units of
Torque
Because we are forced into observing the group behavior, we have the
mathematic equivalent of the simple picture on the right
Resonance--A general
phenomenon for energy pumping
Imagine a kid on a swing…
The period (frequency of the swing is determined (g, r(length), ).
Lets say the natural period is 3 seconds, or the frequency is 0.33
If the Daddy gives a push every 3 seconds, the kid will go higher and energy will be absorbed.
Every 2 seconds and the motion will get stalled and “interfered” with.
Every 1.5 seconds and the energy will get absorbed but not as efficiently. The Daddy will get tired.
This general principal applies in NMR among other kinds of measurement, and holds whether we scan through the applied frequency or multiplex
all at once
NMR-What is it Good For?
(absolutely everything!)
• Solving structures of compounds like synthetics, impurities,
natural products
• Identifying metabolites
• Stereochemical determination
• Follow reactions
• Validating electronic theory; trends within series of compds.
• Kinetics
• Extended structure, e.g. protein nmr
• Molecular interactions e.g. ligand binding
• Acid-base questions
• Purities
• Mechanisms, e.g. isotope distributions, other effects
• Questions about the solid state
• Imaging
And Besides that…
• You get your sample back!
• Not so for mass spec
• Try recovering your compound from a KBr
pellet or nujol mull
But on the other Hand…
• NMR is one of the least sensitive analytical
methods
• Characterized by long relaxation time constants,
limiting experimental efficiency in real time
• Sometimes too much information. Can be
demanding on interpretation skill
• Relatively Expensive compared with other
analytical methods
• As with other methods NMR has “blind spots” and
cannot serve as an analytical panacea
What Do I Hope you will Learn?
• Enough theory to make you conversant in the area….
• NMR with respect to how the effects arise and can be predicted;
connection with experiments and limitations of these; survey of how
the instruments work.
• Basis of the experiments
• Data processing considerations, at level to appreciate what may have
been done to give your result.
• A basic toolbox of experiments, what they do and how to use them in
your work
• A working knowledge of organic chemical shifts and influence of
symmetry on signal counting
• Spin coupling, coupling networks and connectivity, use of J-coupling
constants in chemistry
Why NMR?
•
•
•
•
•
•
Unmatched versatility as an Analytical technique
High on chemical information content
Significant interpretability
Interpretable at several levels of sophistication
Response related to molar preponderance
These attributes are true for solids, liquids,
mixtures, and to a small extent, gas phase
• More than half the periodic table has at least one
NMR active isotope
What are the Measurables in
NMR?
• Intensity (analytical parameter, proportional
to molarity)
• Chemical Shift (the electronic
surroundings)
• Couplings (scalar J and dipolar D; bond
paths, angles connectivity and distances)
• Relaxation parameters (motions, distances)
How do we Generate, and Record
NMR Spectra?
•Pulser
•Frequency generation
•Power Amplifier
•Oscillator
Acquisition
computer
Timing control signals
transmitter
RF pulse
signal
Probe in
Magnet
NMR Acquisition commands
Host Workstation
Network
•User interface
•Expt. Setup, control
•Data processing, plotting
receiver
FID with 90deg phase shift
Data file storage
Free Induction Decay
signal
PreAmp
•Superheterodyne (beat-down to AF)
•Phase sensitive detection
•A/D convertor
Block Diagram for Spectrometer
Radio Frequency TransmitReceive system
•Finely controlled RF pulses
•Microsecond control
•Precise control of timing, e.g pulses
and delays
•Other precisely delivered RF for
decoupling, selective excitation
•Gradient amp and generator,
shielded in probe to avoid eddy
currents
Modern Superconducting NMR Magnets
Older Magnets (1970s)
had opposed pole faces.
High voltages and
currents demanded heroic
temperature control. Field
ran side to side through
sample
Supercon magnets have
much larger fields, better
homogeneity. Field runs up
the axis of the sample.
Note: Special
superconducting
alloys NiobiumTantalum. Search
goes on for higher
temperature
superconductors.
New technology! Built in
auxilliary magnet with
reversed current acts as
“active” shield, partly
eliminating the projection
into the room.
Lines of force project several feet into the room. They concentrate at the top and
bottom. Magnets can grab iron objects and accelerate them.
What’s the role of the magnet?
energy
• Bigger the field strength, the better. This is both
from a sensitivity and dispersion of signals point
of view.
• Expressed in Hz, permits easiser math and trig as
needed. Gauss would generate energies in ergs.
E= h  H0Iz
Remember the
energy difference
gives the population
excess. Roughly
H7/4 increase
Field strength, H0
The NMR Probe
Matching to Tx network
Sample goes inside here
Coil
Tuned
Circuit
Usually there is a double
tuned response for
Deuterium lock
A second coil provides a
decoupling, gradient or
other RF
How Sensitive is NMR?
N 2h 2 I ( I + 1)
H0 = c0H0
=
3 IkT
N=#spins
The Rider site, referenced below gives receptivity vs. 13C with
clickable entries. These reflect natural abundance, , etc.
http://arrhenius.rider.edu/nmr/NMR_tutor/periodic_table/nmr_pt_frameset.html
Another good site is http://nmr.magnet.fsu.edu/resources/nuclei/table.htm
The answer to that question is…
• Not all that Sensitive!
• At any given time Mass spec is at least a
104 times more sensitive
• Compare with UV, IR at least 102x
sensitive
• This is tied to the fact that NMR detects

only the tiny Boltzmann excess. Any old
molecule can fragment in MS or absorb a
IR photon. Lots of research in NMR
aimed at the sensitivity problem
E
N
= e kT
N
k is Boltzmann constant
1.38x10-23J/molecule•K
Because the E is so
small, the excess (which
is what we detect) is
miniscule
,  are the short names
for the upper, lower spin
quantum states of a
spin=1/2 nucleus
Most Important Nuclei in NMR
•1H, (also 2H, 3H)
•13C
•31P
•15N especially when labeled into proteins
•19F
•29Si
•Some isotopes of Sn, Cd. Pb, Ag, Pt
No coincidence that these are the I=1/2 nuclei. Spin
numbers higher possess nuclear quadrupole moment
as well. This couples to, broadens and complicates
the nuclear spin angular momentum. For the most
part these are niche nuclei. Exception is 11B
Quadrupolar Nuclei
• Spin ≤ 1
• Electric field at nucleus non-symmetrical
• Effective relaxation mechanism, promotes loss of
NMR fine structure
• “decouple” from attached spins. Can even wipe
out attached spin 1/2 signals.
• Lines are broad, very challenging NMR
• 35Cl, 11B, 17O, 14N, 7Li, etc.
• Some redeem themselves, deuterium, 6Li