UC4

Quantum Mechanics in 3D

UC4 contents:

• Schrodinger equation in Spherical Coordinates

• Coulomb potential and quantum description of the Hydrogen atom

• Angular momentum and spin

• Larmor precession and the Stern-Gerlach experiment

Schrödinger equation in 3D

The generalisation to three dimensions is straightforward. Schrödinger’s equation says:

the Hamiltonian operator Ĥ is obtained from the classical energy:

by the standard prescription (applied now to y and z, as well as x):

Or
Schrödinger equation in 3D
Thus:

where:

is the Laplacian, in cartesian coordinates.

The potential energy V and the wave function Ψ are now functions of r = (x, y, z) and t.
The probability of finding the particle in the infinitesimal volume d 3r = dx dy dz is
| Ψ(r, t) |2 d 3r, and the normalisation condition reads:

with the integral taken over all space.

Schrödinger equation in 3D
If V is independent of time, there will be a complete set of stationary states,

where the spatial wave function ψn satisfies the time-independent Schrödinger equation:

The general solution to the (time-dependent) Schrödinger equation is:

With the cn constants determined by the initial wave function, Ψ(r,0), in the usual way.

Notice that if the potential admits continuum states, then the sum becomes an integral.
Schrödinger equation in spherical coordinates:
Most of the applications we will encounter involve central potentials, for which V is a
function only of the distance from the origin, V(r) → V(r).
In that case it is natural to adopt spherical coordinates, (r, θ, ϕ) (Figure 4.1).
In spherical coordinates the Laplacian takes the form:
Schrödinger equation in spherical coordinates:
In spherical coordinates, then, the time-independent Schrödinger equation reads

Variable separation:
Schrödinger equation in spherical coordinates:

The Radial Equation

The Angular Equation

The Angular Equation

Multiplying by:

Separation of variables:

Dividing by:
Solution for Φ:

where m can be positive or negative.

when Φ advances by 2π, we return to the same point in space.

m must be an integer:
 Solution for ϴ:

The solution reads:

where Plm is the associated Legendre function, defined by:

for m ≥ 0

and Pl(x) is the lth Legendre polynomial, defined by the Rodrigues formula:
Solution for ϴ:
For negative values of m:

Pℓ(x) is a polynomial (of degree ℓ) in x, and is even or odd according to the parity of ℓ.
Pℓm(x) is not, in general, a polynomial — if m is odd it carries a factor of (1 − x 2)0.5
ℓ must be a non-negative integer.
If m > ℓ, Pℓm(x) = 0.
For any given ℓ, then, there are (2ℓ + 1) possible values of m:
Solution for ϴ:
For negative values of m:

First Legendre polynomials:

Solution for ϴ:
We need Pℓm(cos θ), and (1 − cos2 θ)0.5 = sin θ, so Pℓm(cos θ) is always a
polynomial in cos θ, multiplied (if m is odd) by sin θ .
Normalisation condition: solution for ϴ:
The volume element in spherical coordinates:

and Ω is the solid angle.

Normalisation condition:

It is convenient to normalise R and Y separately:

Normalisation condition: solution for ϴ:
The normalised angular wave functions are called spherical harmonics:

They are orthogonal:

Spherical Harmonics:
 The Radial Equation

The angular part of the wave function, Y(θ, ϕ) is the same for all spherically symmetric
potentials.
The actual shape of the potential, V(r), affects only the radial part of the wave function, R(r).

Variable change:

We get the radial equation:

The Radial Equation
We get the radial equation:

It is identical in form to the one-dimensional Schrödinger, where:

is the effective potential, which contains a centrifugal term:

It tends to throw the particle outward (away from the origin), just like the centrifugal
(pseudo-)force in classical mechanics.

The normalisation conditions is:

which is potential V(r) specific.

 The hydrogen atom

The hydrogen atom consists of proton of charge e,

together with a much lighter electron charge −e.

mp = 1.67 × 10−27 kg
e = 1.60 × 10−19 c
me = 9.11 × 10−31 kg

1 e2
From Coulomb’s law: F = the potential in SI units is:
4πϵ0 r 2

And the radial equation becomes:

Effective potential (Veff)

 The hydrogen atom
We need to solve this equation for u(r), and determine the allowed energies.

The Coulomb potential admits:

- Scattering states (E > 0) -> electron-proton scattering
- Bound states (E < 0) -> hydrogen atom

The Radial Wave Function

We are interested in finding bound states (E < 0) of:

Let’s divide this equation by E and define:

 The hydrogen atom
The Radial Wave Function

We introduce:

Let’s analyse the asymptotic behaviour of this equation.

blows up

(for large ρ)

blows up

(for small ρ)
 The hydrogen atom
The Radial Wave Function
We introduce the new function v(ρ):

(for large ρ)

(for small ρ)

Therefore:

We assume the solution, v(ρ), can be expressed as a power series in ρ:

 The hydrogen atom
The Radial Wave Function
We assume the solution, v(ρ), can be expressed as a power series in ρ, for which we
need to determine the coefficients (c0 , c1 , c2 , . . .).

Replacing into the radial equation above, we get:

 The hydrogen atom
The Radial Wave Function

This recursion formula determines the coefficients, and hence the function v(ρ).

For large j (this corresponds to large ρ, where the higher powers dominate):

If this were the exact result, it blows up at large ρ (so it is not normalisable):
 The hydrogen atom
The Radial Wave Function: the Bohr radius

Thus, the series must terminate:

which makes v(ρ) a polynomial of order (N − 1), with (therefore) N − 1 roots, and
hence the radial wave function has N − 1 nodes.

Let’s define:

Remember:

This is the so-called Bohr radius.

 The hydrogen atom
The Radial Wave Function: the Bohr formula
Remember:

This is the Bohr formula

Therefore:

The spatial wave functions are labeled by three quantum numbers (n, l, and m):

n ≣ principal quantum number

l ≣ azimuthal quantum number
where: m ≣ magnetic quantum number

and v(ρ) is a polynomial of degree n − l − 1 in ρ, whose coefficients are determined (up

to an overall normalisation factor) by the recursion formula:
 The hydrogen atom
Energy levels

Bohr formula:

The ground state (that is, the state of lowest energy) is the case n = 1:

The binding energy of hydrogen (the amount of energy we would have to impart to the
electron in its ground state in order to ionise the atom) is 13.6 eV.

Remember:

Thus, l = 0 and m = 0:
 The hydrogen atom
Energy levels for hydrogen
For arbitrary n, the possible
values of l are:

For each l there are (2l + 1)

possible values of m.

The total degeneracy

of the energy level En is:
 The hydrogen atom
Radial wave function of hydrogen
The polynomial v(ρ) is a function well known to applied mathematicians; apart from
normalisation, it can be written as:

where:

is an associated Laguerre polynomial, and:

is the q-th Laguerre polynomial.

 The hydrogen atom
Laguerre Associated Laguerre
polynomials polynomials
 The hydrogen atom
Radial wave functions for hydrogen:
 The hydrogen atom
Normalised hydrogen wave functions:

The stationary states of the hydrogen atom are labeled by three quantum numbers:
n, l, and m.

The wave functions are mutually orthogonal:

This follows from the orthogonality of the spherical harmonics:

and (for n ≠ n′) from the fact that they are eigenfunctions of Ĥ with distinct eigenvalues.
 The hydrogen atom
Normalised hydrogen wave functions:

They can be visualised via density plots, in which the brightness of the cloud is
2
proportional to ψ .
| |
 The hydrogen atom

The quantum numbers n, ℓ, and m can be identified from the nodes of the wave function.

The number of radial nodes is given by N − 1 (for hydrogen this is n − ℓ − 1).

For each radial node the wave function vanishes on a sphere.

m counts the number of nodes of the real (or imaginary) part of the wave function in the ϕ
direction. These nodes are planes containing the z axis on which the real or imaginary
part of ψ vanishes.

ℓ − m gives the number of nodes in the θ direction. These are cones about the z axis on
which ψ vanishes.
 The hydrogen atom
Or via surfaces of constant probability density:
 The hydrogen atom
The spectrum of Hydrogen:
If we put a hydrogen atom into some stationary state ψnlm, it should stay there forever.

If we tickle it slightly (by collision with another atom, say, or by shining light on it), the
atom may undergo a transition to some other stationary state:
- by absorbing energy, and moving up to a higher-energy state, or
- by giving off energy (typically in the form of electromagnetic radiation), and moving
down.

Such perturbations are always present.

Transitions (quantum jumps) are constantly occurring.
A container of hydrogen gives off light (photons), whose energy corresponds to the
difference in energy between the initial and final states:
 The hydrogen atom
The spectrum of Hydrogen:

According to the Planck formula: and we know:

Rydberg formula

where:

is known as the Rydberg constant (Bohr calculated it!).

 The spectrum of Hydrogen:
Lyman series: transitions to the ground state (nf = 1) lie in the ultraviolet.
Balmer series: transitions to the first excited state (nf = 2) lie in the optical.
Paschen series: transitions to the second excited state (nf = 3) lie in the infrared.

At room temperature, most hydrogen

atoms are in the ground state.
To obtain the emission spectrum we
must populate the excited states.
This is done by passing an electric
spark through the gas.
 The hydrogen atom
Angular momentum:
The principal quantum number (n) determines
the energy of the state:

l and m are related to the orbital angular momentum:

In the classical theory of central forces, energy and angular momentum are the fundamental
conserved quantities. The angular momentum of a particle (with respect to the origin) is
given by:

The corresponding quantum operators are obtained by the standard prescription:

We need to obtain the eigenvalues and the eigenfunctions of the angular momentum
operators.
 The hydrogen atom
Angular momentum: Eigenvalues
The operators Lx and Ly do not commute:

x with px, y with py, and z with pz fail to commute:

We can get the others by cyclic permutation of the indices (x → y , y → z, z → x):

These are the fundamental commutation relations for angular momentum.

Notice that Lx , Ly, and Lz are incompatible observables.

 The hydrogen atom
Angular momentum: Eigenvalues
According to the generalised uncertainty principle:

There are no states that are simultaneously eigenfunctions of Lx and Ly.

The square of the total angular momentum:

does commute with Lx:

 The hydrogen atom
Angular momentum: Eigenvalues

L 2 also commutes with Ly and Lz:

We can hope to find simultaneous eigenstates of L 2 and (say) Lz:

 The hydrogen atom
Angular momentum: Ladder operator technique
Let:

Its commutator with Lz is:

Remember:

Then, f is a common eigenfunction:

Therefore, L± f is also an eigenfunction of L 2 with the same eigenvalue λ.

 The hydrogen atom
Angular momentum: Ladder operator technique
L± f is also an eigenfunction of L 2 with the same eigenvalue λ.

so L± f is an eigenfunction of Lz with the new eigenvalue μ ± ℏ.

L+ is the raising operator: it increases the eigenvalue of LZ by ℏ.

L− is the lowering operator: it lowers the eigenvalue by ℏ.

For a given value of λ, then, we obtain a “ladder” of states, with

each “rung” separated from its neighbours by one unit of ℏ in
the eigenvalue of Lz.
 The hydrogen atom
Angular momentum: Ladder operator technique
There must exist a “top rung”, ft , such that:

Let ℏℓ be the eigenvalue of Lz at the top rung:

Now,

We have:

Thus:
 The hydrogen atom
Angular momentum: Ladder operator technique
There must also exist a “bottom rung”, fb , such that:

Let ℏℓ be the eigenvalue of Lz at the bottom rung:

Remember:

Comparing with:

So the eigenvalues of Lz are ℏm, where m goes from −l to +l, in N integer steps.

It follows that l = -l+N , and hence l = N/2, so l must be an integer or a half-integer.

 The hydrogen atom
Angular momentum: Eigenvalues
The eigenfunctions are characterised by the numbers l and m:

where:

For a given value of l, there are 2l+1 different values

of m (i.e. 2l+1 “rungs” on the “ladder”).

Arrows are possible angular momenta (in units of ℏ),

they all have the same length.
Their z components are the allowed values of m
(-2,-1, 0, 1, 2).
 The hydrogen atom
Angular momentum: Eigenvalues

Arrows are possible angular momenta (in units of ℏ),

they all have the same length:
Their z components are the allowed values of m
(-2,-1, 0, 1, 2).

The magnitude of the vectors (the radius of the

sphere) is greater than the maximum z component:

The uncertainty principle implies that we cannot know all three components of L.

Actually, there aren’t three components — a particle simply cannot have a determinate
angular momentum vector.
If Lz has a well-defined value, then Lx and Ly do not.
 The hydrogen atom
Angular momentum: Eigenfunctions
We will see that f ℓm = Yℓm, i.e. the eigenfunctions of L 2 and Lz are the spherical harmonics.

Let’s rewrite Lx, Ly, and Lz in spherical coordinates:

Since:

Here:

The unit vectors θ ̂ and ϕ̂ can be resolved into their cartesian components:
 The hydrogen atom
Angular momentum: Eigenfunctions

In components:

We also need the raising and lowering operators:

And:
 The hydrogen atom
Angular momentum: Eigenfunctions

In particular:

We are now in a position to determine f ℓm(θ, ϕ).

It is an eigenfunction of L 2, with eigenvalue ℏ2ℓ(ℓ + 1).

But this is precisely the “angular equation”:

 The hydrogen atom
Angular momentum: Eigenfunctions
And it’s also an eigenfunction of Lz, with the eigenvalue ℏm:

which is equivalent to the azimuthal equation:

We have already solved this system of equations!

The result (appropriately normalised) is the spherical harmonic, Yℓm(θ, ϕ).

Conclusion:
Spherical harmonics are the eigenfunctions of L 2 and Lz. When we solved the Schrödinger
equation by separation of variables, we were inadvertently constructing simultaneous
eigenfunctions of the three commuting operators H, L 2 and Lz:
 The hydrogen atom
Angular momentum: Eigenfunctions
We can rewrite Schrödinger’s equation as follows:

The algebraic theory of angular momentum permits ℓ (and hence also m) to take on
half -integer values:

Separation of variables yielded eigenfunctions only for integer values.

Are the half-integer solutions spurious?

No, they are of profound importance, as we shall see next.

 The hydrogen atom
Spin:
In classical mechanics, a rigid object admits two kinds of angular momentum:
orbital ( L = r × p ), associated with motion of the center of mass, and
spin ( S = I ω ), associated with motion about the center of mass.

In quantum mechanics, the distinction is absolutely fundamental.

Orbital angular momentum, associated (in the case of hydrogen) with the motion of the
electron around the nucleus (and described by the spherical harmonics).
Spin, which has nothing to do with motion in space (and not described by any function
of the position variables r, θ, ϕ) but which is somewhat analogous to classical spin.

The electron (as far as we know) is a structureless point, and its spin angular momentum
cannot be decomposed into orbital angular momenta of constituent parts.

Suffice it to say that elementary particles carry intrinsic angular momentum (S) in
addition to their “extrinsic” angular momentum (L).
 The hydrogen atom
Spin:

Fundamental commutation relations:

It follows (as before) that the eigenvectors of S 2 and Sz satisfy:

where S± ≡ Sx ± iSy. The eigenvectors are not spherical harmonics (they’re not
functions of θ and ϕ at all), and there is no reason to exclude the half-integer
values of s and m:
 The hydrogen atom
Spin:

Every elementary particle has a specific and immutable value of s, which we call
the spin of that particular species:

- π mesons have spin 0

- electrons have spin 1/2
- photons have spin 1
- Δ baryons have spin 3/2
- gravitons have spin 2; and so on.

By contrast, the orbital angular momentum quantum number ℓ (e.g. for an electron
in a hydrogen atom) can take on any (integer) value, and will change from one to
another when the system is perturbed.
s is fixed for any given particle, so the theory of spin is comparatively simple.
 The hydrogen atom
Spin 1/2:
s = 1/2 is the spin of the particles that make up ordinary matter (protons, neutrons,
and electrons), as well as all quarks and all leptons.

There are just two eigenstates:

1. spin up (informally, ↑):

2. spin down (informally, ↓):

Using these as basis vectors, the general state of a spin-1/2 particle can be represented by
a two-element column matrix (or spinor):

where:
 The hydrogen atom
Spin 1/2:
The spin operators become matrices:

If we write S2 as a matrix with undetermined elements:

Conclusion:
 The hydrogen atom
Spin 1/2:
Similarly:

For which:

Remember:
 The hydrogen atom
Spin 1/2 (Pauli spin matrices).

These are the famous Pauli spin matrices.

Sx, Sy, Sz, and S 2 are all hermitian matrices (as they should be, since they represent
observables).
On the other hand, S+ and S− are not hermitian—evidently they are not observable.

The eigenspinors of Sz are:

 The hydrogen atom
Spin 1/2 (Pauli spin matrices).
Remember:

If you measure Sz on a particle in the general state χ, you could get:

+ℏ/2, with probability | a |2 ,or

-ℏ/2, with probability | b |2 .
Since these are the only possibilities: | a |2 + | b |2 = 1
(i.e. the spinor must be normalised: χ † χ = 1.

But what if, instead, we chose to measure Sx?

What are the possible results and probabilities?

We need to know the eigenvalues and eigenspinors of Sx.

The characteristic equation is:
 The hydrogen atom
Spin 1/2 (Pauli spin matrices).
We need to know the eigenvalues and eigenspinors of Sx
The characteristic equation is:

The possible values for Sx are the same as those for Sz. The eigenspinors are obtained via:

The (normalised) eigenspinors of Sx are:

 The hydrogen atom
Spin 1/2 (Pauli spin matrices).
The (normalised) eigenspinors of Sx are:

The generic spinor χ can be expressed as a linear combination of them:

If you measure Sx:

ℏ 1
The probability of getting: + is | a + b |2 ,
2 2
ℏ 1 2
and the probability of getting: − is |a − b|
2 2
 The hydrogen atom
Spin 1/2 (implications):

For a particle in the state χ+, what is the z-component of a particle’s spin angular
momentum?
We can answer unambiguously: +ℏ/2.

What is the x-component of that particle’s spin angular momentum?

If we measure Sx, the chances are fifty-fifty of getting either +ℏ/2 or -ℏ/2.

It simply does not have a particular x-component of spin.

 The hydrogen atom
Addition of Angular Momenta
Suppose now that we have two particles, with spins s1 and s2.

we denote the composite state by:

What is the total angular momentum of the system?

 The hydrogen atom
Addition of Angular Momenta
What is the total angular momentum of the system?

What is the net spin, s, of the combination, and what is the z component, m?

The z component is easy:

Thus:

The net spin, s, is much more subtle. If you combine spin s1 with spin s2, what total
spins s can we get?
The answer is that you get every spin from (s1 + s2) down to (s1 − s2)

or (s2 − s1), if s2 > s1 in integer steps:

1
 The hydrogen atom
Addition of Angular Momenta
Roughly speaking, the highest total spin occurs when the individual spins are aligned
parallel to one another, and the lowest occurs when they are antiparallel.

Example 1:
If you package together a particle of spin 3/2 with a particle of spin 2,
you could get a total spin of 7/2, 5/2, 3/2, or 1/2, depending on the configuration.

Example 2:
If a hydrogen atom is in the state ψnlm , the net angular momentum of the electron (spin
1 1
plus orbital) is ℓ + or ℓ −
2 2
If you now throw in spin of the proton, the atom’s total angular momentum quantum
number is ℓ + 1 or ℓ − 1.
Notice that ℓ can be achieved in two distinct ways, depending on whether the electron
1 1
alone is in the ℓ + configuration or the ℓ − configuration.
2 2
 The hydrogen atom
Addition of Angular Momenta
The combined state | s m ⟩ with total spin s and z-component m will be some linear
combination of the composite states | s1 s2 m1 m2 ⟩:

Because the z components add, the only composite states that contribute are those for
which:

The constants: are called Clebsch-Gordan coefficients.

Example 1: The shaded column of the 2 × 1

If two particles (of spin 2 and spin 1) are at rest in a box, and the total spin is 3, and its z
component is 0, then a measurement of Sz could return the value ℏ (with probability 1/5),
(1)

or 0 (with probability 3/5), or −ℏ (with probability 1/5). Notice that the probabilities add up
to 1 (the sum of the squares of any column on the Clebsch–Gordan table is 1).
 The hydrogen atom
Clebsch-Gordan coefficients
 Electron in a Magnetic Field

A spinning charged particle constitutes a magnetic dipole. Its magnetic dipole moment,
μ, is proportional to its spin angular momentum, S.

The proportionality constant, γ, is called the gyromagnetic ratio.

The gyromagnetic ratio of an object whose charge and mass are identically distributed is
gs q
, where gs is the spin g-factor, q is the charge and m is the mass.
2m
For reasons that are fully explained only in relativistic quantum theory, the gyromagnetic
e
ratio of the electron is (almost) exactly twice the classical value: γ = −
m

When a magnetic dipole is placed in a magnetic field B, it experiences a torque, μ × B,

which tends to line it up parallel to the field (just like a compass needle).
 Electron in a Magnetic Field

The energy associated with this torque is:

so the Hamiltonian matrix for a spinning charged particle, at rest in a magnetic field B, is:

where S is the appropriate spin matrix:

in the case of spin 1/2.

 Electron in a Magnetic Field

Larmor precession
Imagine a particle of spin 1/2 at rest in a uniform magnetic field, which points in the z-
direction:

The Hamiltonian is:

The eigenstates of H are the same as those of Sz :

Larmor precession
The energy is lowest when the dipole moment is parallel to the field—just as it would be
classically.

Since the Hamiltonian is time independent, the general solution to the time-dependent
Schrödinger equation,

can be expressed in terms of the stationary states:

The constants a and b are determined by the initial conditions:

where | a |2 + | b |2 = 1.
Larmor precession
With no essential loss of generality we can write a = cos(α/2) and b = sin(α/2),
where α is a fixed angle whose physical significance will appear in a moment.
Thus :

Let’s calculate the expectation value of S, as a function of time:

Larmor precession

Similarly:

Thus, ⟨S⟩ is tilted at a constant angle

α to the z axis, and precesses about
the field at the Larmor frequency:

just as it would classically. The Ehrenfest

theorem guarantees that ⟨S⟩ evolves
according to the classical laws.
Stern-Gerlach experiment
In an inhomogeneous magnetic field, there is not only a torque, but also a force, on a
magnetic dipole:

This force can be used to separate out particles with a particular spin orientation.
Imagine a beam of heavy neutral atoms, traveling in the y direction, which passes
through a region of static but inhomogeneous magnetic field (Figure 4.15)—say

where B0 is a strong uniform field and the constant α describes a small deviation from
homogeneity.
Stern-Gerlach experiment
The force on these atoms is:

But because of the Larmor precession about B0, Sx oscillates rapidly, and
averages to zero; the net force is in the z direction:

and the beam is deflected up or down, in proportion to the z component of the spin
angular momentum.

Classically we’d expect a smear (because Sz would not be quantised), but in fact the
beam splits into 2s + 1 separate streams, demonstrating the quantisation of
angular momentum.
In silver atoms all the inner electrons are paired, in such a way that their angular
momenta cancel. The net spin is simply that of the outermost— unpaired—electron, so
1
in this case s = , and the beam splits in two.)
2
Stern-Gerlach experiment

The Stern–Gerlach experiment has played an important role in the philosophy of

QM, where it serves both as the prototype for the preparation of a quantum state
and as an illuminating model for a certain kind of quantum measurement.

We tend casually to assume that the initial state of a system is known (the
Schrödinger equation tells us how it subsequently evolves)—but it is natural to
wonder how we get a system into a particular state in the first place.

If we want to prepare a beam of atoms in a given spin configuration, we pass an

unpolarised beam through a Stern–Gerlach magnet, and select the outgoing
stream we are interested in (closing off the others with suitable baffles and shutters).

Conversely, if we want to measure the z component of an atom’s spin, we send it

through a Stern–Gerlach apparatus, and record which bin it lands in. This may not
be the most practical way to do the job, but it is conceptually very clean, and hence
a useful context in which to explore the problems of state preparation and
measurement.
Stern-Gerlach experiments ?

Experiment 1:

When a second, identical, S-G apparatus is placed at the exit of the first apparatus, only
z+ is seen in the output of the second apparatus.

This result is expected since all neutrons at this point are expected to have z+ spin, as
only the z+ beam from the first apparatus entered the second apparatus

Source:
https://en.wikipedia.org/wiki/Stern%E2%80%93Gerlach_experiment
Stern-Gerlach experiments ?

Experiment 2:

The middle system shows what happens when a different S-G apparatus is placed at the
exit of the z+ beam resulting of the first apparatus, the second apparatus measuring the
deflection of the beams on the x axis instead of the z axis.

The second apparatus produces x+ and x- outputs. Now classically we would expect
to have one beam with the x characteristic oriented + and the z characteristic oriented +,
and another with the x characteristic oriented - and the z characteristic oriented +.
Stern-Gerlach experiments ?

Experiment 3:

The output of the third apparatus which measures the deflection on the z axis again
shows an output of z- as well as z+.
Given that the input to the second S-G apparatus consisted only of z+, it can be inferred
that a S-G apparatus must be altering the states of the particles that pass through it.

This experiment can be interpreted to exhibit the uncertainty principle:

Since the angular momentum cannot be measured on two perpendicular directions at the
same time, the measurement of the angular momentum on the x direction destroys
the previous determination of the angular momentum in the z direction.
That's why the third apparatus measures renewed z+ and z- beams like the x
measurement really made a clean slate of the z+ output.
Stern-Gerlach experiment

Source: https://www.youtube.com/watch?v=PH1FbkLVJU4