Basics of Quantum Mechanics: James Binney Oxford University

Basics of Quantum Mechanics
James Binney
Oxford University
• The book
Available at Clarendon Reception for £20
Also for free download at
http://www-thphys.physics.ox.ac.uk/
people/JamesBinney/QBhome.htm
• The film: podcasts can be reached from
http://www-thphys.physics.ox.ac.uk/people/JamesBinney/lectures.html
Physics
• It’s about predicting the future from knowledge of the present
• We do it with numbers
• Knowledge of the present derives from measurements
• Measurements are prone to error – our knowledge is imperfect
) physics is ultimately probabilistic
– eg ladder
– eg pendulum
• To push physics to its limits you must quote probabilities
– eg R=14 § 0.1 Ohms
Measurement 1
• To measure you must disturb
• The disturbance may be too small to matter
– measure a star’s position!
• But often the disturbance matters
– eg measuring V across a circuit component
• Small things are more strongly disturbed by measuring
kit than large ones
• Atoms, electrons, etc are significantly disturbed
• Ideal measurements are reproducible:
– if I say “the momentum p of this electron is 3 GeV/c” I’m
claiming that if you measure p with precision, you’ll get 3
GeV/c
Measurement 2
• Key to QM is the idea that any system has states in which
the outcome of a measurement is certain – these states are
abstractions but crucial abstractions
– eg |E1> is state in which a measurement of energy will yield E1 J
– eg |+> is a state in which a measurement of the z-component of
spin angular momentum will yield +½~ (kg m2/s)
– eg |E1+> is a state in which the results of measuring either E and
sz are certain
– eg |p> is a state in which a precision measurement of
momentum is certain to yield p GeV/c
• In a generic state |Ã>, the result of measuring E is uncertain
• But after a high-precision measurement the result of
measuring E again is certain (reproducibility!)
• So the act of measuring E jogs the system from the generic
state |Ã> into one of the special states |Ei>
Measurement 3
• If we do a high-precision measurement of p when the
system is in the state |Ã> we jog it into a state |pi> in which
the result of measuring p again is certain
• In general a precision measurement of E when the system
is in the state |pi> yields an uncertain result – we can only
calculate probabilities Pji of finding Ej
• Once we have found Ej and jogged the system into the state
|Ej> the result of measuring p is uncertain because the
system is no longer in one of the special states in which the
outcome of a precision p measurement is certain
• That is, each thing you can measure jogs the system into
one of a different set of states, so it’s not possible to get
the system into a state in which the outcome of any
precision measurement is certain
– measurements are generally incompatible
– dynamical variables are questions you can ask, not intrinsic
properties
Quantum physics
• We take on board that
– we have to calculate probability distributions P(x) not just
expectation values <x>
– measurements disturb the system & leave it in a state that
differs from the pre-measurement state
• Q physics tackles these tasks using the idealisation of
reproducible measurements
• So far everything has been straightforward & inevitable
– this is just grown-up physics
• But it’s clear that Q physics is going to be
mathematically more challenging than C physics
because calculating a whole (non-negative) function
P(x) is much harder than calculating one number <x>
Quantum amplitudes
• Q physics is built on a wonderful mystery:
– It (& it alone) obtains a probability P from a complex number A
the quantum amplitude for P:
– P=|A|2
• Nobody knows why this is the correct thing to do
• No application of this formalism has been successful
outside Q physics
• The whole mathematical formalism of Q physics follows
naturally & easily once you accept the use of quantum
amplitudes
• The formalism is immensely convenient
– It allows us to calculate probability distributions much more
easily than in C physics
• Aren’t we lucky: in our hour of need a powerful new
formalism comes to our rescue!
Quantum interference
• Quantum amplitudes have a key, logic-defying
property:
– If something can happen in 2 mutually exclusive ways,
1 and 2, and the amplitude for it to happen by route 1
is A1 and by route 2 is A2 then the probability for it to
happen by either 1 or 2 is
P1+2 = |A1+2|2 = |A1+A2|2 =|A1|2 + |A2|2 +(A1*A2+A1A2*)
= P1 +P2 + 2Re(A1*A2)
• That is: we add amplitudes not probabilities
• The extra term is a manifestation of “quantum
interference”
2-slit interference
• Expect |A1|2 to be roughly Gaussian

• Write
• Near centre line p1(x) ' p2(x) and P(x)

oscillates from 0 to 4P1(x)
Quantum states 1
• There are certain things we can measure
• “observables” – a terrible name
• With each observable Q there is a list of possible values
qi returned by a precise measurement of Q
• The set of qi is called the spectrum of Q
– eg spectrum of x coordinate is (-1 ,1)
– eg spectrum of KE is (0,1)
– eg spectrum of any component of angular momentum is {…,
(k-1)~, k~, (k+1)~,..), where k=0 or ½ and ~ = 1.05 £ 10-34 J s
• Elements of the spectrum are called allowed values of Q
Quantum states 2
• With each element of the spectrum qi there is a probability
amplitude Ai that a precise measurement will return that value and
a state |qi> in which the system will be left after the measurement
• QM is the science of calculating from the set {Ai} the amplitudes,
say {aj}, for getting a value such as bj on measuring another
observable B
• A complete set of amplitudes contains sufficient amplitudes to
enable the amplitudes for any measurement to be predicted
• Conventionally a complete set is a minimal set:
– None of its members can be calculated from a knowledge of the other
members alone
• A complete set of amplitudes characterises the current state of the
system as precisely as is physically possible
• That state, |Ã>, is pointed to by the complex numbers {Ai} in just
the way a geometric point a is pointed to by its coordinates (ax,ay,az)
– So |Ã> $ {Ai} just as a $ {ai}
• |Ã> is a vector with complex components
Quantum states 3
• Just as many different sets of coordinates (ax,ay,az) or
(ar,aµ,aÁ) all pick out the same geometrical point a, so many
sets of amplitudes pick out the same physical state |Ã>
• By designating a state |Ã> (“ket psi”) we keep open our
options as to which complete set of amplitudes we will use
for calculations
• In C physics choosing the appropriate coordinate system is
often the key to solving a given problem
• In Q physics choosing the appropriate set of amplitudes is
often the key
– eg we can specify the state |Ã> of an electron by giving the
amplitudes a(p) to measure momentum p or the amplitudes
Ã(x) to measure location x
– Ã(x) is called the wavefunction and its values are quantum
amplitudes
Dirac notation 1
• We already discussed the physical significance of the sum
of 2 amplitudes
• So if |Ã >=(A1,A2,..) and |Á >=(B1,B2,..) are 2 states of the
same system, we should consider
– |Ã>+|Á> $ (A1+B1, A2+B2,..)
– Standard rule for adding vectors
• Because probabilities for all possibilities must sum to 1, we
require i |Ai|2=1 and i |Bi|2=1, & we need to normalise
|Ã>+|Á> by multiplying by ® = 1/(i|Ai + Bi|2)1/2
• So a new physical state is|Ã’> = ®(|Ã>+|Á>)
• Objects that you can add & multiply by numbers constitute
a vector space
• It’s often useful to choose a basis {|i>} for a vector space:
• Any state |Ã>=i ai|i> for some amplitudes ai
Dirac notation 2
• With every vector space V we get the dual space V’ for free:
– V’ is the space of all linear (complex-valued) functions on V
• We denote members of V’ by <f| (“bra f”) & then <f|Ã> is a
(complex) number, the value taken by the linear function
<f| on the vector |Ã>
– In traditional notation f(|Ã>)
• If |i> is a basis for V, a basis for V’ is provided by the
functions <j| defined by the rule
– <j|i> = ±ij
• Given |Ã> = i ai|i> we choose to define
– <Ã| = j aj*<j| so that
– <Ã|Ã> = ij aj*ai<j|i> = i|ai|2 = 1
• If |Á> = j bj|j> then
– <Á|Ã> = i bi*ai = (<Ã|Á>)*
Energy representation
• For a particle trapped in a potential well the
spectrum of energy E is discrete so there are
states |Ei> in which a measurement of E has a
certain outcome
• These states form a basis for V so any state
– |Ã> = i Ai |Ei>
• If we “bra through” by <Ej| we have
– <Ej|Ã> = Aj
• This is a key rule & explains the importance of
bras: they enable us to extract experimentally
important amplitudes from the system’s state

Basics of Quantum Mechanics: James Binney Oxford University

Uploaded by

Copyright:

Available Formats

Basics of Quantum Mechanics: James Binney Oxford University

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Basics of Quantum Mechanics: James Binney Oxford University

Uploaded by

Copyright:

Available Formats

Basics of Quantum Mechanics

• Expect |A1|2 to be roughly Gaussian

• Near centre line p1(x) ' p2(x) and P(x)

You might also like