Hydrogen Atom Radial Equation

Mayukh Mazumdar
Assistant Professor, Department of Physics
Salesian College (Autonomous) Siliguri
October 2023

1 Setting up the Equation

So, now that we have already segregated the Schrodinger’s equation for the
hydrogen atom into a radial and an angular equations, its time to solve the
previous one. Let us rewrite the radial equation now:

d 2 dR 2mr 2
dr (r dr ) − ℏ2 [V (r) − E]R = l(l + 1)R.....[1]

To simplify the above equation, let us perform a change of variable:

u(r) = rR(r)

Now, let us see how the terms in [1] change due to this. We will now have

R = ur
=⇒ dR 1 du u
dr = r dr − r 2
2 dR du
=⇒ r dr = r dr − u.....[2]

Thus, putting [2] in [1], we will get

d du 2mr 2 u l(l+1)u
dr (r dr − u) − ℏ2 [V (r) − E] r = r
2 l(l+1)u
=⇒ r ddru2 +  du
−
dr
du
 − 2mr
dr ℏ2 [V (r) − E]u = r
2 l(l+1)u
=⇒ r ddru2 − 2muV ℏ 2 + 2mEu
ℏ 2 = r
ℏ2 d2 u ℏ2 l(l+1)u
=⇒ − 2m dr 2 + V u + 2m r 2 = Eu

We will now put the explicit form of the potential function V (r) for the hy-
drogen atom:

2
e
V (r) = − 4πϵ0r

1
Thus, the radial equation finally takes the following shape:

ℏ2 e2 ℏ2 l(l + 1)
− + [− + ]u = Eu .....[3]
2m 4πϵ0 r 2m r2

So, our target is to solve [3] and obtain the radial wave function R(r).

2 Solving the Radial Equation [3]

In order to solve [3], we will be performing a number of substitutions in the
process. Hence, the reader is asked to remain cautious and not get foxed by the
number of new characters that will be following soon!
To start things off, let us call a new character κ:
√ 2 2
−2mE κ
κ= ℏ .....[4] ⇒ E = − ℏ2m
It must be noted that κ is a real quantity as E < 0 for the hydrogen atom (a
bound state).
Putting [4] in [3], we get:

E d2 u e2 u E l(l+1)u
κ2 dr 2 − 4πϵ0 r − κ2 (κr)2 = Eu

Further simplifications will lead to the following form:

1 d2 u me2 1 l(l+1)
κ2 dr 2 = [1 − 2πϵ0 ℏ2 κ (κr) + (κr)2 ]u.....[5]

Here we can find repetitive terms (κr and its exponent) and also note a clumsy
me2
constant term ( 2πϵ 2 ). Hence we seek to simplify [5] further via:
0ℏ κ

me2
ρ = κr, ρ0 = 2πϵ0 ℏ2 κ .....[6]

This calls for revisiting the terms in [5], especially the term on the L.H.S.

du dρ
= du
dr
du
dρ dr = κ dρ
One further differentiation leads to our desired quantity:

d2 u d du d du dρ d 2

dr 2 = dr ( dr ) = dρ ( dr ) dr = κ dρ (κ du 2d u
dρ ) = κ dρ2
1

Therefore, by [5], we get:

2 ρ0 l(l+1)
1 
κ2 ddρu2 = [1 − ρ + ρ2 ]u
κ2


d2 u ρ0 l(l+1
=⇒ dρ2 = [1 − ρ + ρ2 ]u.....[7]

1 Remember! κ is a constant quantity.

2
 We will now examine the forms of the solutions of [7] for extreme values of
ρ:
(a) As ρ → ∞, [7] approximates to

d2 u
dρ2 ≈u

The general solution of the above is

u(ρ) = Ae−ρ + Beρ ,where A,B are arbitrary constants.

However, the acceptable form of the solution will be only the first term [Check
yourself why]. Hence,
u(ρ) = Ae−ρ .....[8], for very large ρ.

(b) As ρ → 0, the centrif ugal term (= l(l+1)u

ρ2 ) dominates over the other
2
terms in the R.H.S of [7]. Thus [7] becomes:
d2 u l(l+1)u
dρ2 ≈ ρ2

The general solution of the above is of the form

u(ρ) = Cρl+1 + Dρ−l .....[9]3

However, the acceptable solution will be

u(ρ) = Cρl+1 .....[10], for very small ρ.
So, we can say that the solution u(ρ) will be a combination of both this extreme
solutions (infact, at least a product of the two). To seek for the complete form
of the solution, we now move a step further and try to bid good-bye to the
asymptotic term (e−ρ ). For this, we perform the following transformation:
u(ρ) = ρl+1 e−ρ v(ρ).....[11]
New challenge unlocked! Find v(ρ).

l −ρ
du
dρ = (l + 1)ρ e v − ρl+1 e−ρ v + ρl+1 e−ρ dv
dρ
2
d u l−1 −ρ l −ρ
⇒ dρ 2 = l(l + 1)ρ e v − (l + 1)ρ e v+.......work
youself! [Hint: there will
be nine terms.]
Once you do the hard work and put the results in [7], you will get the equation
in terms of v now:

d2 v dv
ρ 2
+ 2(l + 1 − ρ) + [ρ0 − 2(l + 1)]v = 0 ....[12]
dρ dρ
2 See Appendix 2
3 See Appendix 1

3
As you might have become annoyed with too many substitutions and started
cursing me (?), let me assure you that there is no need of further substitutions.
All we need to do now is to solve [12]. We will use the best tool from our kit
for this - the F robenius method.
The solution can be expressed as an infinite series

P∞ j
v(ρ) = j=0 cj ρ

Differentiating, we will get

d2 v
dv
P∞ P∞ P∞
dρ = j=0 jcj ρj−1 = j=0 (j + 1)cj+1 ρj and, dρ2 = j=0 j(j + 1)cj+1 ρj−1

Putting these in [12] a simplifying a bit, we get the recursion relation

2(j + l + 1) − ρ0
cj+1 = cj .....[13]
(j + 1)(j + 2l + 2)

Let us now see how the coefficients (c′j s) look like for large values of j. For
large j, [13] becomes

2j 2
cj+1 ≈ j(j+1) cj = j+1 cj

Assuming the above relation to be exact, we can look into some cj values:

If j = 0,

c1 = 2c0 ,

c2 = 22 c1 = 2.2
2.1 c0 ,

c3 = 23 c2 = 2.2.2
3.2.1 c0 ,

......this is enough for us to generalise the recursion relation as

2j
cj = c0
j!

j 2j j
= c0 e2ρ
P P
Hence, v(ρ) = j cj ρ = c0 j j! ρ

By [11], we get that

u(ρ) = c0 ρl+1 eρ .....[14]

4
The overall behaviour of the solution has asymptotic nature (eρ )! Despite the
best of our efforts, and abiding by Physics at every step, we could not avoid the
asymptotic nature at the end. However the situation is not so hopeless. Since
our formalism has physically acceptable ingredients, we can still justify our ef-
forts. This can be done if we notice that the series solution has to truncate at
some maximum value of j, say jmax :

cjmax+1 = 0 ⇒ 2(jmax + l + 1) − ρ0 = 0 ⇒ ρ0 = 2n, where n = jmax + l + 1 .

So, finally we have reached the stage to find the radial wavefunction
(at least the ground state one) [I hope you haven’t forgotten this radial
wavefunction amidst all the substitution ruckus!]. Recall the total hydrogen
wavefunction was

ψ(r, θ, ϕ) = Rnl (r)Ylm (θ, ϕ)

and now we can write that

Rnl = u(r)
r = rρ
1 l+1 −ρ
e v(ρ)
2(j+l+1−n)
and, cj+1 = (j+1)(j+2l+2) cj

For the ground state of the hydrogen atom, n = 1, l = 0, m = 0.

Thus the ground state radial wavefunction will look like

R10 (r) = 1r ρl+1 e−ρ v(ρ)|n=1,l=0 = 1r ρe−ρ c0 4

It is time to replace the ρ in a meaningful way. For this,let us recall [6]

me2 me2 1
κ= 2πϵ0 ℏ2 ρ0 = 4πϵ0 ℏ2 n = an
2
4πϵ0 ℏ
where a = me2 ≈ 0.53Å.......the Bohr radius.

r r
Also, ρ = κr = an ⇒ ρ=
an
Thus we revert to the ground state radial wavefunction
r
c0 − a
R10 (r) = ae . Normalisation will give us the value of the constant c0 .

Normalising R10 (r):

R∞ 2
|R10 (r) |r2 dr = 1 ⇒ c0 = √2
0 a

4 The recursion formula truncates after the first term for n=1. So, v(ρ) = c0

5
 Also, Y00 = √14π . So, the ground state wavefunction of the hydrogen
atom is given by

1 r
ψ100 (r, θ, ϕ) = √ e− a .....[15]
πa3

Congratulations! You did it. Now take a small break before we try to retrieve
some other important physical quantities related to the hydrogen atom.

3 Energy Eigen Values

Let us recall [4]:

2 2
κ
E = − ℏ2m
Now, this can be re-written as
2 2 4 4
E = − 2m
ℏ m e
4π 2 ϵ20 ℏ4 (2n)2
= − 32π2me
ϵ20 ℏ2 n2
= −E
n2 , where E1 is the ground state
1

energy.
If you are willing to run your fingers on the calculator (and ofcourse put the
correct values of the constants!), you will get E1 = −13.6eV .
Thus, we get the Bohr’s formula:

E1 5
En = , where n = 1, 2, 3, ......
n2
Thus, we got the expression for the energy eigen values for the hydrogen atom.

4 Excited States of Hydrogen Atom

Before you already start building your castle of high hopes(!), let me break your
heart by giving a reality check - we will stick to the First Excited State
only.
For the first excited state, rather states, we have n = 2; l = 0, 1. For
l = 0, m = 0 and for l = 1, m = −1, 0, 1. Hence we used the plural for ′ state′ .

If l = 0, the recursion relation with n = 2 becomes:

2(0+0+1−2)
c1 = (0+1)(0+0+2) c0 = −c0 .
5 It does make sense at this point of time to put the subscript n in E to show that energy

is quantised.

6
Putting j = 1 gives c2 = 0. Thus, v(ρ) = c0 + c1 ρ = c0 (1 − ρ).

Hence

c0 r r
R20 (r) = (1 − )e− 2a , since ρ = r
2a for n = 2.
2a 2a
Similarly, for l = 1, the recursion relation ends after the first term [Try yourself
by putting l = 1, j = 0, n = 2 in the recursion relation].Thus, v(ρ) = c0 in this
case.
Hence
c0 − r
R21 (r) = re 2a
4a2
You can try your luck with finding the expression for R2−1 (r) yourself now. I
feel I have given enough motivation!

5 Appendix 1
We have the following differential equation:

d2 u
= l(l+1)u
dρ2 ρ2 .....[1], where l is some real constant.
Let us try to seek a solution of the above equation. Evidently, its solution must
be a function that yields the constants in increment of 1 (l and l + 1). Hence
it can be an algebraic function whose exponent is a function of this constant.
Mathematically speaking, let the trial solution be

u(ρ) = Cρl ,where C is an arbitrary constant. Successive differentiation gives us

d2 u l(l−1) l l(l−1) l(l+1)

dρ2 = ρ2 Cρ = ρ2 u ̸= ρ2 u.

So, our type of function is fine, except the fact that its exponent should be
modified:

u(ρ) ρl+q , where q is a constant and qϵR. For our case, q = 1. Now you
can repeat the previous calculations with u(ρ) = Cρl+1 and find

d2 u
dρ2 = l(l+1)
ρ2 Cρ
l+1
= l(l+1)
ρ2 u, which matches with the R.H.S of [1]. So, we
got one probable solution for [1]; there will be another one as well since it is
a second order differential equation. However, it will not be very hard to seek
that u(ρ) = Dρ−l can be another possible solution. Hence, the general form of
the solution of [1] will be:

u(ρ) = Cρl+1 + Dρ−l , where C and D are arbitrary constants.

7
6 Appendix 2
Recall that we got the differential equation in Appendix 1 by claiming that the
centrifugal term had dominated the remaining terms. Here, I have plotted two
functions on the Desmos graphing calculator FOSS- the red one represents the
mathematical nature of the L.H.S of [7] while the purple one represents the
same for the centrifugal term. It can be clearly seen that for very small values
of ρ, both the curves nearly overlap.

Fig.1: Visualizing the Dominance of the Centrifugal Term in [7]

Thus we can justify our claim regarding the dominance of centrifugal term.

7 References
Griffiths, D.J. (2018). Introduction to Quantum Mechanics