This paper is based on the following assumptions:

1) Finite mass of the universe.

2) Newton’s absolute time and space.

Then, the following two conditions:

− Equality between the speed of light c and the total escape speed u.

− Light composed of longitudinal-extended, elastic particles, as defined on Section 1.4.

We obtain, on the 1^st Part, these main results:

1) The measured constancy of c is due, first, to the constancy of the total gravitational potential U on Earth: indeed, the max variation of U on Earth, occurring between Aphelion and Perihelion, leads to a negligible Δu_AP = 0.10 m∙s^-1; moreover, the speed of light emitted by a source, either fixed or in motion from its initial location O (where the photons emission starts), is invariant with respect to O;

2) Doppler effect equations for the light, slightly different from the relativistic ones;

3) The compensating velocity, (to restore the resonance source-absorber on Harvard Tower Experiment), has same value but contrary direction with respect to the one predicted by GR;

4) The variation of U from ground to GPS satellite level, implying a decrease of c and consequently a decrease of the frequency of the photons emitted by atomic clocks at that altitude, give them an increase of their counted time (inducing them to run faster);

5) High gravitational redshifts are due to the increase of c (inducing the increase of the photons length λ) during their path toward the Earth where |U_o| >> |U_→∞|.

On the 2^nd part, under the third condition claiming that the electron charge can be considered as a point-particle fixed to the electron surface, and facing the atom nucleus during the electron orbit, we got these results:

On H atom:

6) n = 1, 2, ∙∙∙, 137 electron circular progressive orbits, not an infinite number; n also represents the number of admitted photons (of the same ray) absorbed (or emitted) by the electron during two electron orbits: (in fact, the photons total number along each circular double-orbit (in short d-orbit) becomes n², as shown on Section 2.4);

7) The electron radius becomes r e = [ ( 1 / 137 α ) − 1 ] r 0 (instead of its classical value r_e = α²r_B with r B = α / 4 π R ∞ the Bohr radius and α the fine structure constant); r 0 = α / 4 π R H is the orbit of the electron charge, while the orbit of the electron barycenter (electron effective orbit) becomes r ′ 0 = r 0 + r e = r 0 / 137 α ;

8) The electron charge orbital speed along r₀ becomes υ 0 = α c , while the electron effective orbital speed along the orbit r ′ 0 becomes υ ′ 0 = c / 137 ;

9) The electron radial speed (due to the impacts photons-electron) becomes w n 2 = c/137² constant for every circular orbit; along the orbit #137, the electron orbital speed becomes υ 137 = c / 137 2 and since w n 2 = υ 137 there is ionization.

10) On Photoelectric effect the number of photons hitting a circling electron varies from n_f related to the specific threshold frequency n_f (=W_f/h) to (n_f)^½ related to a higher frequency corresponding to the lower frequency occurring for the Compton effect; for instance, it is shown that, as for Caesium (having W_f @ 2 eV), n_f = 361;

11) Compton effect (CE) needs one photon only; our Doppler effect equations for the light lead to Compton Eq., while, as for the Relativity, the CE is not a Doppler effect.

As known, considering in space only one mass M, (for simplicity its barycenter), the gravitational potential U in a point O, assuming U_¥ = 0, is U = −MG/s, with s the distance M-O. The conservation of energy applied to a particle m (of unitary mass) coming from the infinity, gives E = U + K, with K = ½u² its kinetic energy, therefore for E = 0, one finds U = −K, yielding

u = − 2 U = 2 M G / s (1)

which is a scalar, (called escape speed), representing the value of the velocity u, (referred to M), a massive particle needs to reach the infinity; hence the relation

v Mm = u [with | u | = u = ( − 2 U ) 1 2 ] (2)

represents the condition, as for a point-particle m, to be provided with the absolute escape velocity u (absolute means here referred to M) whereas considering a generic point O,

v Om ≡ u O [with | u O | = u O = ( − 2 U O ) 1 2 ] (3)

represents the condition, regarding m, to be provided with the relative escape velocity u_O, where relative means here referred to a generic point O; (anyhow, our claimed variable length of photons, during their emission from a source S gives them an invariant speed referred to any considered point O, in spite of any velocity v_OS).

Regarding the possibility to define the escape speed from two or more masses, we point out that considering M as a real mass, the potential U, in a considered point, has to be regarded as the sum of the partial potentials U_n due to each partial mass M_n composing the mass M; therefore, as for two masses, we have

U 1 , 2 ≡ U 1 + U 2 = − ( M 1 G / s 1 ) − ( M 2 G / s 2 ) (4)

yielding u 1 , 2 ≡ − 2 U 1 , 2 = − 2 ( U 1 + U 2 ) thus we can write

u 1 , 2 2 = − 2 U 1 , 2 = u 1 2 + u 2 2 (5)

while for all the n masses in space we get

u ≡ ( − 2 U ) ½ = ( − 2 Σ U n ) ½ = ( 2 Σ M n G / s n ) ½ = Σ u n 2 (6)

representing the total escape speed.

For instance, the escape speed from the Earth surface, due to Earth and Sun only, becomes u E , S 2 = u E 2 + u S 2 = ( 2 M E G / R E ) + ( 2 M S G / d ) with R_E the Earth radius, and d the distance Earth-Sun giving u_E,S ≅ 42 km/s, while the escape speed from our Galaxy can be roughly expressed as u_g ≃ (2M_gG/d_g)^½ where M_g ≃ 10⁴² kg is our Galaxy mass and d_g ≃ 30 kly ≃ 3 ´ 10²⁰ m the distance between the Earth and our Galaxy centre, giving u_g ≃ 8 ´ 10⁵ m∙s^-1.

Now, the mass of universe, by some authors, is estimated [1] [2] [3] to be M_u ≅ 10⁵³ kg; about the same value is given through the number ( ≅ 10²²) of observable stars [4] [5] [6] , and since from Earth the distribution of the masses appears to be homogeneous and isotropic, under our assumption U_¥ = 0, we may assume their density as decreasing toward the infinity like a function ρ = ρ_ce^−as being ρ_c ≅ 9.2 ´ 10⁻²⁷ kg/m³ the critical density.

Writing the mass of universe as

M u = ∫ 0 ∞ 4 π s 2 ρ c e − a s d s = 4 π ρ c ∫ 0 ∞ s 2 e − a s d s = 8 π ρ c a 3 ≅ 10 53 kg (7)

we find

a = ( 8 π ρ c / M u ) 1 3 ≅ 1.3 × 10 − 26 m − 1 (8)

and since, on Earth, the variation of potential due to an increase of the distance ds, can be written dU = −dm G/s with dm = ρ4πs²ds and ρ = ρ_ce^−as, the potential on Earth becomes

U 0 = − ∫ 0 ∞ ( 4 π s 2 / s ) G ρ c e − a s d s = − 4 π ρ c G ∫ 0 ∞ e − a s s d s = − 4 π ρ c G / a 2 ≅ − 4.5 × 10 16 J (9)

yielding

u 0 = ( − 2 U 0 ) ½ ≅ ( 9 × 10 16 ) ½ ≅ 3 × 10 8 m / s . (10)

We can therefore infer that, on Earth, u₀ = c₀, giving c 0 2 / 2 = − U 0 and, in general,

c = ( − 2 U ) ½ = u . (11)

The equality c = u, which implies the massiveness of light, means that, along any free path, the speed of light only depends on the value of the potential along that path.

The equality c = u is also supported by a cosmological reason: in fact, if c > u the energy of light will be lost forever and furthermore the observable masses, following the always increasing mass of light going toward the infinity, will also tend to the infinity moving away from each other. On the contrary, if c < u, all the masses in space, (having speed lower than u), will tend to a gravitational collapse, whereas for c = u, the mass of light, tending to the infinity in an unlimited time, appears to be the necessary mass to avoid the two events (collapse or dispersion).

On Earth, a small variation of the total escape speed u_o, from Equation (11) written as u² = −2U yielding 2u = −2dU/du, can be written as

Δ u = − Δ U / u 0 (12)

where ΔU is the variation of the total potential on Earth, mainly due to the variable distance Earth-Sun; so, between Aphelion and Perihelion, where a = (1 + e)d and p = (1 − e)d are the respective distances Earth-Sun, with e (=0.0167) the eccentricity of the Earth’s orbit and d ≅ 1.5 ´ 10¹¹ m the average distance Earth-Sun, and since pa ≅ d², and (a - p) = 2ed, we get

Δ u AP = − Δ U AP / u 0 = − ( U P − U A ) / u 0 = [ ( M S G / p ) − ( M S G / a ) ] / u 0 ≅ M S G ( a − p ) / d 2 u 0 ≅ M S G 2 e d / d 2 u 0 ≅ M S G e / d u 0 ≅ + 0.10   m s − 1 (13)

with M_S = 2 ´ 10³⁰ kg the mass of the Sun, and we note that Δu_AP is compatible with the accuracy of the value of the speed of light in vacuum (on Earth) c₀ = 299,792,458 m∙s⁻¹.

The Galileo’s velocity composition law (since different parts of a moving object may have different velocity) is strictly valid for point-particles, and therefore the light cannot be a succession of point-particles (otherwise it should follow this Galileo’s law), hence we may infer that the light, to comply both this law and the equality c = u, could be composed of particular particles, photons, (whose physical characteristics are then shown on §2.2), as defined:

Longitudinally-extended, elastic (variable length) and massive particles emitted by a source during an emission time T at speed c = u moving individually or along rays, (continuous succession of photons where every tail corresponds to the front of the next one).

Photons speed: the measurement of the speed of a point-particle (or regarding an object, its barycentre), is defined through the ratio v = d/t with d the distance between two Observers (O₁ and O₂) and t (=t₂ - t₁) the time the particle/barycentre needs to cover such a distance; on the contrary, every photon may have different length with respect to two Observers, (like, for instance, an elastic thread during its stretching), hence to define the speed of one photon, we must consider its length λ divided its transit time T_t (time the whole particle needs to cross one Observer). Therefore the relation

c = λ / T t (14)

represents the speed of one photon referred to a considered Observer, whereas the relation c = d/t, written as

c = d / t = n λ / n T t = λ / T t (14-a)

where n is the number of photons (of one ray) each of them having length λ, and where t = nT_t is the transit time of the n photons, represents the average speed of the light along d.

Referring to Figure 1, let O be the location (with respect to all the masses in

space) of the source S at the time (t = 0) the front A is emitted; therefore O represents the location of the initial emission of a considered photon.

The Earth’s surface, where c₀ is practically constant, has to be intended as the basic reference frame for the parameters of light (λ, ν) while other frames, because of the Doppler effect, would give, for the same source, different values. So, still referring to Figure 1, we intend the frame O as fixed to the Earth’s surface. We also intend:

T: emission time of the considered photon AB.

λ: length of the emitting photon if S is fixed to O during T.

λ_O: length of an emitting photon if S during T is in motion with respect to O.

T_t : transit time of one photon (T_O for O, T_P for P, while T_S = T ).

On Figure 1, if S during T is fixed to O, the speed of the front A (considered as a point-particle), see Equation (3), is υ_OA = u_O, where u_O only depends on the location of O, thus, at t = T, the length λ of the photon AB corresponds to the path covered by A, so we have

υ OA T = u O T ≡ (= υ SA T for S fixed to O during T). (15)

If S has velocity v_OS with respect to O, the velocity of the front A, referred to S, being v_SA = v_SO + v_OA and since from (3), v_OA = u_O, becomes

v SA = v SO + u O ⇒ υ SA = u O ± υ OS (16)

with the sign + if photon and source have opposite direction; then

υ SA T ( ≡ λ O ) = u O T ± υ OS T ⇒ λ O = λ ( 1 ± β ) [with β = υ OS / u O ] (17)

where λ_O is the length of the emitted photons when S is in motion from O. We point out that the variation of the photons length only depends on v OS .

Now, the transit time T_t for the Observer O, can be written

T O = T ± υ OS T / u O = T ( 1 ± β ) [with β = υ OS / u O ] (18)

corresponding to the algebraic sum of the time T (=λ/u_O) the front A needs to cover the path λ, plus the time the tail B (emitted, like every part of the photon, at speed u_O referred to O), needs to cover the path S_T - O = υ_OST, that is ΔT = υ_OST/u_O. (As for the observer P showed on Figure 1, being v_OP = 0, then T_P = T_O).

Thus, the speed of the photon AB referred to O, since T_O increases/decreases together with λ_O, as per Equation (14), becomes

u O = λ O T O = λ ( 1 ± β ) T ( 1 ± β ) = λ T ≡ c (19)

showing that the speed of a photon referred to an Observer fixed to the source initial location, (O), is invariant despite any motion of the source with respect to O.

The frequency of the light referred to a generic Observer O, on our bases, has to be defined as ν = n/T_t with n the number of photons, of the same ray, crossing the Observer during their transit time T_t . Hence, for T_t = 1 s, we have ν = n meaning that the frequency corresponds to the number of photons, same ray, transiting in 1 s, while, for n = 1, the frequency becomes

ν = 1 / T t (20)

We point out that the photons frequency emitted by a source S, under the condition υ_OS = 0, has to be equal to the one observed by O; this is also valid if O and S belong to different potential, (e.g., the equality of the number of balls falling from the top of a tower with respect to an Observer at the tower base), and therefore the condition υ_OS = 0 always implies n_S = n_O whatever is the distance source-observer.

[We also point out that, according to the Relativity, (due to the claimed constancy of c and because of the Doppler effect), an Observer in relative motion from a source should observe different wavelength as well as different frequency for emitted light; on the contrary, on our results, if the source is fixed to its initial location (the ground on Earth), the length of the emitted photons is invariant, while c and ν are varying, as hereafter shown].

Referring now to Figure 2, let P be an observer (represented at different times, by P_A and P_B) moving with velocity v_OP from O along the same direction of the photons while the source S is moving from O with velocity v_OS contrary to them; we assume the photon AB totally emitted at the time the front A reaches P_A (for simplicity as t = 0). [In fact, after this emssion, the source may even disappear].

Now, the transit time T_P of the photon AB is given, see Equation (18), by T_O plus the time T_OP (= ± υ OP T / c ) due to the motion of P from O, and since v_SP = v_SO + v_OP, we can write

T p = T ± υ SO T c ± υ OP T c ( = T ± υ SP T c ) = T ( 1 ± β ± β P ) ,     [ with   β p = ± υ OP c ] (21)

corresponding to the interval between the time the front A reaches P_A (at t = 0) and the time the tail B leaves P_B; hence

c p = λ O / T p = λ ( 1 ± β ) / T ( 1 ± β ± β p ) = c ( 1 ± β ) / ( 1 ± β ± β p ) . (22)

[In short, the motion of S (from O) causes a variation of the photon length, while the relative motion observer-source varies the observer transit time and its related frequency].

On next Figure 3, we analyse the parameters c, λ, ν of five configurations regarding a source S, the considered frame O (practically the Earth’s surface) and an Observer P.

where

β = υ_OS/c, relative speed of S with respect to O,

β_p = υ_OP/c relative speed of P,

c = λ/T basic speed of light, referred to conf. #1,

λ_O = λ(1 ± β) length of emitted photon,

T_S = T photon transit time for S, equal to its emission time,

T_O = T(1 ± β); T_t for O; sign + if S and photon have opposite direction,

T_P = T(1 ± β ± β_p); T_t for P; +β_p if P and photon have same direction,

c_S = λ_O/T_S = λ_O/T; speed of light observed by S,

c_P = λ_O/T_P speed of light observed by P,

c_O = λ_O/T_O speed of light observed by O.

Regarding the measurements of c through the method c = d/t carried out along a double path, the light emitted inside the measurement system (MS), because of the interaction light-matter, has to be absorbed and then re-emitted by the 2^nd Observer which, during the re-emission, becomes a new emitting source; therefore the above equations regarding transit time and photon length, are also valid for the re-emitted photons which now have contrary direction, hence β' = −β and also λ' = −λ_O giving to c, along a double path, the constant value c.

[Light emitted outside the MS: at the time the light is crossing the MS, this becomes the new emitting source].

The Equation (17) represents our longitudinal Doppler effect for the light, while the general case of this effect, assuming υ_OS << λ (hence OP ≅ PB), see Figure 4(a), can be written as

λ 0 = λ ± υ OS T cos α = λ ( 1 ± β cos α ) (with β = υ OS / c ) (23)

with α the angle between the direction of λ_O and v_OS.

As for the Transverse Doppler effect, see Figure 4(b), in general, we have

λ 0 = λ 2 ± ( υ OS T ) 2 = λ 1 ± β 2 (24)

As for a source circling around an Observer P, the Figure 5 shows that the length of the photon λ emitted when its source is fixed to O, becomes λ_O when S is circling with velocity v_OS.

[The succession of the photons λ represents a ray if S is at rest with P, while the succession λ₁ ® λ₃ represents a ray if S is circling around P, with O, B, C the emission points of the circling source]. Then, assuming λ_O > λ, we have

λ O = λ 2 + ( υ OS T ) 2 = λ 1 + β 2 = λ 1 + ω 2 r 2 / c 2 (25)

while the photon transit time, being T the emission time of λ becomes

T O = T 1 + β 2 = T 1 + ω 2 r 2 / c 2 (26)

with r the orbit radius, ω the source angular speed, giving to any photon, speed c_O = c.

General Relativity predicts that the gravitational field of the Earth will cause a photon emitted downwards (towards the Earth) to be blueshifted: scope of HTE experiment was to detect this shift of light. A Mossbauer source S was placed on the top of a tower (height h @ 22 m) emitting photons toward the tower base where an Absorber went out of resonance; the experiment did not clarify if the non-resonance was due to a variation of λ or ν; indeed, because of the claimed constancy of c, both of them should have varied their initial values during the path tower-base.

According to our bases, we have different evaluations of this experiment. Let us first consider S at the base and A at its top, see Figure 6.

It is a fact that if S and A are at relative rest at the same level (e.g. both on the ground), then A is in resonance and it is also a fact that other experiments effected at this regard, e.g. Blatt, [7] , show that if S moves toward the absorber, not contrarily, the absorber goes out of resonance: indeed, if S is fixed to the ground, the length of emitted photons (as previously shown) is constant, whereas if S moves with respect to its initial emission location (the ground), the photon length varies, thus we may infer that the resonance is physically related to the constancy of the length of photons.

Now, referring to Figure 6, if S, on the ground, emits photons reaching A at height h, because of the variation of potential, following our relation c = (−2U)^½, it will be c₀ ® c_h; since S and A are at reciprocal rest, it turns out that, as shown just after the Equation (20), it is always ν_h = ν₀ (in spite of any variation of potential between S and A) yielding λ_h = c_h/ν_h. Therefore, with r₀ the radius of Earth (hence r_h = r₀ + h) and M_E the mass of Earth, writing the (11) as c² = −2U, we get 2c = −2dU/dc hence from r₀ to r_h and since Δc = −ΔU/c with ΔU = U_Eh - U_E0 = − (M_EG/r_h) + M_EG/r₀, we find

c h − c 0 c 0 = − Δ U c 0 2 = − M E G h r h r 0 c 0 2 ≅ − g h / c 0 2 ⇒ c h = c 0 ( 1 − g h / c 0 2 ) (27)

(where ΔU = gh); then for n_h = n₀ we have

λ h = c h / ν h = c 0 ( 1 − g h / c 0 2 ) / ν 0 = λ 0 ( 1 − g h / c 0 2 ) ⇒ ( λ h − λ 0 ) / λ 0 = − g h / c 0 2 = c h − c 0 c 0 (28)

and since λ_h < λ₀, contrary to Relativity, A observes a blue-shift effect.

Thus, to restore the resonance via Doppler Effect (DE), see Figure 6(b), S has to recede from A in order that the photon length could increase, see Equation (17), from λ_h to λ 0 = λ h ( 1 + β ) with β = υ AS / c ≡ υ / c ; therefore, equating the relative variation of the photon length due to the DE, that is (λ_h − λ₀)/λ₀ = −υ/c, to the one due to the altitude, Δλ/λ₀, as given by (28), we get

υ / c = g h / c 0 2 ⇒ υ = g h / c 0 = 7.5 × 10 − 7 m / s (for h = 22.5 m) (29)

i.e. the value of the compensating velocity.

This value is also predicted by GR which, implying a decrease of ν for the light moving from the base to the top, predicts an opposite direction of the compensating velocity v with respect to the one shown on Figure 6(b); at this regard, Pound-Rebka and Pound-Snider, [8] [9] [10] , gave no clear indication about the direction of v, since S and A, during the experiment, were moved sinusoidally.

Now, see Figure 7, with A on the base, and taking S to the top, the experience

shows that the absorber goes out of resonance. Indeed, with S on the top, the (27) shows c_h < c₀, but what about the initial parameters of the photons emitted in altitude, n_h and λ_h?

Well, it is known that atomic clocks give an increase of their counted time in altitude (T_h > T₀, hence ν_h < ν₀), therefore taking these clocks from ground to h, their frequency n_h has to follow the same decrease as c_h; this implies λ_h = λ₀ so the frequency of photons emitted in altitude becomes

ν h = c h λ h = c 0 ( 1 − g h c 0 2 ) λ 0 = ν 0 ( 1 − g h c 0 2 ) = ν 0 ( 1 − Δ U / c 0 2 ) . (30)

Now, with S emitting photons toward the base, since S and A are at rest, their frequency remains constant during their path top-base, meaning that ν h − 0 = ν 0 ( 1 − Δ U / c 0 2 ) , while the length of photons reaching the base, and considering that now h is negative, becomes

λ h − 0 = c 0 / ν h − 0 = c 0 / ν 0 ( 1 − Δ U / c 0 2 ) = λ 0 ( 1 + g h / c 0 2 ) (31)

showing an increase of λ from the top to the base. Hence the absorber, on the base, will state, contrary to GR, a red-shift, hence, to compensate it via Doppler shift, see Figure 7, S has now to move toward A; on the contrary, according to GR, A and S should recede from each other. We also highlight two points:

− The increase of λ for photons moving from h to the ground, gives a reason to the gravitational red shift related to far sources.

− The decrease of ν during the path h-ground gives a reason to the lower frequency of the light coming from far sources.

Time dilation: Atomic clocks in altitude (h-clocks) are ticking faster than identical clocks on the ground (g-clocks): at height h, see (27), we have ( c h − c 0 ) / c 0 = − Δ U / c 0 2 , while taking a clock to a GPS satellite, see also Figure 7(a), from (30), one finds

T h = T 0 / ( 1 − Δ U c 0 2 ) = T 0 / ( 1 − δ ) ⇒ T h − δ T h = T 0 [where Δ U c 0 2 ≡ δ ] (32)

with T_h the counted time of one photon emitted by a h-clock, while T₀ by a g-clock. Thus the variation of the counted time between the two clocks, for every emitted photon, due to the altitude, becomes

Δ T ph = T h − T 0 = T h δ (33)

and since the frequency of photons is their number emitted in 1 s along one ray, the term

n 1 s ≡ ν 0 t 1 s (with t 1 s = 1   s ) (34)

is the number of photons (atomic transition of Cs 137) constituting one-second on the ground; then, since ν h = ν 0 ( 1 − Δ U / c 0 2 ) , see (30), the following relation

Δ n 1 s = ν h Δ T ph = ν 0 ( 1 − Δ U c 0 2 ) ( T 0 Δ U c 0 2 ) 1 − Δ U c 0 2 = ν 0 T 0 Δ U c 0 2 = Δ U c 0 2 (35)

is the variation of n 1 s emitted in 1 s by a h-clock; now, see (27), ΔU = M_EGh/r_hr₀ and since r_h ≅ 26,600 km , r₀ ≅ 6400 km, with h ≅ 20,200 km , the increase of counted time in one day ( Δ T 1 d ) of a h-clock, with respect to a g-clock, becomes

Δ T 1 d = Δ n 1 s × 86400   s = 86400 Δ U / c 0 2 = 45.5   μs (36)

hence this effect is a consequence of the variation of U; indeed, as shown later on, the values of the H atom frequencies depend on c, thus, on our bases, on U.

Now let us find the variation of the counted time, between the two clocks, due to their relative motion; thus, being v_S = 2 (2πr_h/86,400 = 3870 m/s the orbital speed of GPS satellites, (corresponding to two orbits/day), it turns out that the counted time variation, between a h-clock and an Observer E at the centre of Earth, due to their relative motion, is given by Equation (26) which, in our case, with β_S = v_S/c (≅1.3 × 10⁻⁵) becomes

T E = T h 1 + β S 2 ≅ T h ( 1 + β s 2 2 ) (valid for β S 2 << 1) (37)

with T_E the photon counted time for the Observer E; then, with v₀ the Earth’s rotational speed (@ 460 m/s at the equator), since β 0 2 = (v₀/c)² << β S 2 = (v_S/c)² and calling T 0 ≅ T h ( 1 + β 0 2 ) the photon counted time of a g-clock with respect to a clock at the Earth’s centre E, we can write T₀ ≅ T_E; thus, the difference between the two transit times T_h and T₀ (≅ T_E) due to their relative motion, is

Δ T ′ ph = T h − T 0 ≅ T h − T E = T h − T h ( 1 + β s 2 2 ) = − T h β S 2 / 2 . (38)

Then, as above, Δ n ′ 1 s ≡ ν h Δ T ′ ph = − ν h T h β S 2 / 2 = − β S 2 / 2 is the variation of the number of photons emitted by a h-clock in 1 s; so the variation of the counted time, in one day, is

Δ T ′ 1 d = − ( β S 2 2 ) 86400   s = − 7.2   μs (39)

showing a decrease of the counted time for a g-clock due to their relative motion; hence

Δ T tot 1 d = Δ T 1 d + Δ T ′ 1 d = 38.3   μs / day (40)

which is also predicted, (with different reason), by GR. To prevent this effect, before launching, the daily counted time (T_1d) of clocks, has to be decreased by ≅38 μs.

In short, the time dilation is a consequence of the variation of U, hence of c: indeed, as show on §2.3, the admitted/emitted frequency on H atom (and of course the one of the other elements) varies together with c.

Gravitational redshift

As for the Relativity, the only way to explain high cosmological redshifts is the Doppler effect, (which implies an incredible universe expansion at a speed υ_u ≅ c), whereas, on our results, disregarding the reciprocal motion between a (far) source and an Observer on Earth, which implies ν = ν₀, we get c/λ = c₀/λ₀, where ν₀, c₀ and λ₀ are the values on Earth, hence for c₀ > c one gets λ₀ > λ, that is a red shift observed on Earth; therefore, the shifts here observed can be expressed as

z ≡ Δ λ λ = Δ c c = c 0 − c c = c 0 c − 1 = U 0 / U − 1 (41)

with U₀ the potential on Earth, U the one on the source. Thus, apart from Doppler effects, z turns out to be the variation of c (as well as λ) during the path of light toward a different potential. In particular, with s the distance Earth-source, for s < @45 Mpc, [11] , (roughly corresponding to −0.01 < z < +0.01) if U (potential on the source) is, in absolute value, higher than the potential on Earth U₀, the (41) gives, on Earth, z < 0 (blue shift), and vice versa for |U| < |U₀|; thus, for s < @45 Mpc, these red/blue shifts indicate that the potential, may increase or decrease with respect to |U₀|. In the range @0.01 < z < @0.20, (where z follows the Hubble’s law), the (41), written as

U = U 0 / ( 1 + z ) 2 ≅ U 0 / ( 1 + 2 z ) ≅ U 0 ( 1 − 2 z ) (valid for z << 1) (42)

shows that, for z << 1, U depends linearly on z, as Hubble’s law; then, for s → ∞, U → 0, hence z → ∞, see Table 1.

[For s > @45 Mpc, z is always positive, hence we may infer that our galaxy is practically near/close to the middle of the masses of universe (where |U| has the max value)].

On our basis, (light composed of our photons), the interaction light-matter requires that to move a circling electron toward outer orbits, the impact photon-electron has to occur, see Figure 8(a), in a radial way, (giving origin to the radial velocity w), otherwise, other impacts could cause the electron fall into the nucleus, due, for instance, to an impact where the velocity of photons and electron have contrary direction.

To be radial, the impact must occur in a specific point, fixed to the electron surface, we call it Impact Point, which, during the electron, orbit, has to face its nucleus, (up to its removal), giving to the electron one rotation every orbit; the impacting photons have to approach the nucleus, as shown on Figure 8(b), perpendicularly to the electron orbit plane, providing, to the electron, a radial velocity w.

Moreover, we can infer that the charge of the electron, has to correspond to the Impact Point (I_p), so we may infer that each photon front is provided with a positive charge, while its tail with an equal negative one. In fact, the photon may be represented as an electric bipole.

The total energy of a mass m is E = m c 2 also proved [12] by the evidence of nuclear reactions like n + p d + γ , hence this energy has to be valid for the (massive)

light too, thus writing their energy as E = 1 2 m c 2 + 1 2 m c 2 we have to infer that photons are provided, out of their kinetic energy K c = 1 2 m c 2 with an internal

energy K i equal to K c and we point out that, toward the infinity, (where U_∞ → 0 hence c_∞ → 0), both K_c and K_i tend to 0; then, since for the light E = hν, the following relation

E = K c + K i = m c 2 = h ν (43)

has to represent the energy of one ray of light, (where photons are flowing), and where

m = h ν / c 2 ≡ γ ν (44)

is the mass of light, having frequency ν, passing along one ray in 1s, while the constant

γ = h / c 2 = m / ν = m T = 7.372495 × 10 − 51 kgs (45)

represents the mass of one photon crossing an observer during the transit time T, while the Planck’s constant

h = γ c 2 (46)

turns out to be the energy of one photon (mass of light of one ray passing during T).

Now, since m is the mass of light passing along one ray in 1s, the term n r m c 2 , with n_r the number of rays emitted by a source S, becomes the energy emitted by S in 1 s; this unitary (unit of time) energy shall be equal to the supplied power P during 1 s, yielding

n r m c 2 = P (47)

therefore

m tot = n r m = P / c 2 (48)

is the total mass lost per second by a source of light; e.g. for a 1 W lamp, we get

m tot = P / c 2 ≅ 1.1 × 10 − 17 kgs − 1 (49)

while the number n_r of rays is

n r ( = m tot / γ ν ) = P / c 2 γ ν = P / h ν (50)

in our case, n_r ≅ 3 ´ 10¹⁸ rays, and we point out that, as for a given power P, the higher is the frequency, the lower is the number of rays, as shown by (50) written as n_rν = P/h. Then, the number of photons n_γ emitted in 1 s becomes

n γ = ( n r ν ) = P ν / h ν = P / h (51)

which, for P = 1 W, gives n_γ = h⁻¹ (=1.5 ´ 10³³ photons/s), thus the inverse of Planck’s constant turns out to be the number of photons emitted in 1s by a source of unitary power, and this great number of photons allows the light to be treated as a wave.

Now, during inelastic impacts photons-electron, (like on absorption or photoelectric effetcs), both kinetic and internal energy of the light are involved, so the momentum transferred to the electron is

p = 2 m c = 2 γ ν c = 2 γ c / T (52)

whose validity is confirmed, see Table 2 of chapter 2.4, by the equality w₁₃₇ = υ₁₃₇ (with υ₁₃₇ the electron orbital speed along the H atom orbit # 137 and where w₁₃₇ is the value of the electron radial speed along such orbit, as a condition for the H atom ionization.

Regarding elastic impacts, the momentum transferred to the electron, is

p ′ = m ′ c = γ ν ′ c = γ c / T ′ (53)

either for incident or for reflectedphotons, with T ′ the total impact time for this interaction; through the (53), and also through our Doppler effect equation, see (17), we got the Compton equation (which cannot be obtained by the Relativity via their Doppler effect equations).

Figure 9(a) represents, on our bases, the H atom basic configuration; on the intermediate conf. (b) the electron is represented circling along the same orbit as its charge, while on conf. (c) the electron reduced mass m r = m e / ( 1 + m e / m P ) = m e / ( 1 + ε m ) , with ε_m = m_e/m_p, circling around the proton mass m_P fixed as origin, is represented either along the effective orbit r ′ 0 or along r₀; referring to conf. (a) we define:

m_e: electron mass;

m_p: proton mass;

υ_e = |v_e| electron ground-state (orbital) speed,

υ_p = |v_P| proton ground-state speed,

r_B the ground-state orbit of the electron charge,

r_e the electron radius,

ε_r ≡ r_e/r_B

r ′ B = ( r B + r e ) = r B ( 1 + ε r ) , ground-state orbit of the electron barycentre,

r_p the proton ground-state orbit.

Now, the constancy of momentum referred to conf. Figure 9(a) gives

m e υ e = m p υ p ⇒ υ p = m e υ e / m p ≡ ε m υ e (54)

then, equating on (a) the frequency of m_e and m_p, we get

υ e / r ′ B = υ p / r p ⇒ r p ( υ p / υ e ) r ′ B = ε m r ′ B = ε m r B ( 1 + ε r ) (55)

The conf. (a), moving m_e (without its charge) from the orbit r ′ B to r B (the electric forces acting on m_e are not involved), turns to conf. (b) where the equality between the frequency of m_e and the one of its charge, gives

υ e / r ′ B = v e / r B ⇒ v e = υ e / ( 1 + ε r ) (56)

then the constancy of momentum referred to conf. (b), m e v e = m p v p , yields

v p = m e v e / m p = ε m v e ⇒ v p / v e = ε m (57)

then, equating on (b) the frequency of m_e and m_p, we get

v e / r B = v p / r ′ p ⇒ r ′ p = ( v p / v e ) r B = ε m r B (58)

and finally, see conf. (c), we find

r 0 = r B + r ′ P = r B + r B ε m = r B ( 1 + ε m ) = r ′ B ( 1 + ε m ) / ( 1 + ε r ) (59)

where m_r is circling with speed υ₀ = v_e + v_p = v_e(1 + ε_m); then

r ′ 0 = r 0 + r e = r 0 + ε r r B = r 0 [ 1 + ε r / ( 1 + ε m ) ] = r 0 ( 1 + ε m + ε r ) / ( 1 + ε m ) (60)

Now, it is known that, on H atom, the admitted frequencies have to satisfy the relation

ν n = ν 0 / n 2       ( n = 1 , 2 , 3 , ⋯ ) (61)

with ν_n the photons admitted frequency along the n-th circular orbit and since, as claimed on Section 1.4, ν (=n/t) is the number of photons (of the same ray) crossing an observer during the time t, the integer n represents the number of photons absorbed by the electron during the photon-electron impact time t and this number, for all the H atom circular orbits, is an integer starting with 1 along the electron ground-state orbit.

The massiveness of light implies a finite impact time during which the electron is circling, and supposing, for now, that, for n = 1, the photon frequency ν₀ should be equal to the electron frequency along the ground-state orbit, that is ν_e0 = υ₀/2πr₀, where r₀, see conf. (c) is the electron ground-state orbit, with υ₀ its orbital speed, we could write

ν 0 / ν e 0 = 1 (first assumption); (62)

to verify this ratio or to find the correct value, we may start from the experimental value of the ¹H ground-state ionization energy W₀ (electron extraction work) corresponding to

W 0 = − U r 0 / 2 = e 2 / 8 π ε 0 r 0 = ½ m r υ 0 2 ≅ 13.8954   eV (63)

with U r 0 = − e 2 / 4 π ε 0 r 0 the potential of the dipole electron-proton, yielding

r 0 = e 2 / W 0 8 π ε 0 ≅ 5.29461 × 10 − 11   m (64)

(rough value of the Bohr radius), thus the orbital speed of the electron along r 0 becomes

υ 0 = e 2 / 4 π ε 0 r 0 m r ≅ 2.18770 × 10 6   m / s (65)

Then, if W₀ is supplied by one ray of light (with energy E = hν₀) it must be

½ m r υ 0 2 = h υ 0 ⇒ ν 0 = m r υ 0 2 / 2 h ≅ 3.28808 × 10 15 s − 1 (66)

with ν 0 the admitted frequency along r₀; hence, the ratio ν₀/ν_e0 becomes

υ 0 / υ e0 = ( T e0 / T 0 ) = υ 0 2 π r 0 / υ 0 = 0.499998 (67)

meaning that during one electron orbit it should transit half of a photon, so we may infer that along r₀ it should be

2 T e0 / T 0 = 2 ν 0 / ν e0 = 1 exact (68)

meaning that, along r 0 the impact time T 0 lasts for two electron circular orbits.

Now, see conf. (c) equating along r₀ the electron centrifugal force to the Coulomb force we have

m r υ 0 2 / r 0 = e 2 4 π ε 0 r 0 2 ⇒ m r υ 0 2 = e 2 / 4 π ε 0 r 0 = − U r 0 (69)

which, for a circular orbit, and since, see (66), m r υ 2 = 2 h ν , gives

ν n = e 2 / 4 π ε 0 r n 2 h (70)

then

r n = e 2 / 8 π ε 0 h ν n = e 2 n 2 / 8 π ε 0 h ν 0 = r 0 n 2 (71)

while as for the orbital speed of the electron m_r along any circular orbit r n the (65) gives

υ n 2 = e 2 / 4 π ε 0 r n m r = e 2 / 4 π ε 0 r 0 n 2 m r = υ 0 2 / n 2 ⇒ υ n = υ 0 / n (72)

and finally the electron frequency along a circular orbit becomes

ν e n = υ n / 2 π r n = υ 0 / n 2 π r 0 n 2 = ν e 0 / n 3 . (73)

Now, along r₀, the (70) yields

ν 0 = e 2 / 4 π ε 0 r 0 2 h = e 2 c / 2 ε 0 h c 4 π r 0 = α c / 4 π r 0 (74)

with α = e²/2ε₀hc the fine structure constant and given 2 ν 0 / ν e0 = 1 (exact) we have

δ 1 ≡ 2 ν 0 / ν e0 = ( 2 α c / 4 π r 0 ) / ( υ 0 / 2 π r 0 ) = a c / υ 0 = 1 (75)

yielding

υ 0 = α c = e 2 / 2 h ε 0 = 2187691.2   m / s (76)

and then the (66) becomes

υ 0 = m r υ 0 2 / 2 h = m r α 2 c 2 / 2 h = R H c (77)

where R H = m r α 2 c / 2 h = 1 / λ 0 is the Rydberg constant.

Now, from (61) and (73), because of the ratio δ 1 ≡ 2 ν 0 / ν e0 = 1 , we also get

δ n ≡ 2 ν n / ν e n = ( 2 ν 0 / n 2 ) / ( ν e 0 / n 3 ) = n     ( with   n = 1 , 2 , 3 , ⋯ ) (78)

which written as nT_n = 2T_en shows that the impact time nT_n of n photons (with frequency ν_n) equals the time the electron needs for two orbits along any circular orbit r_n.

Now the correct value of r₀, from (71), becomes

r 0 = e 2 8 π ε 0 h ν 0 = e 2 8 π ε 0 h R H c = α 4 π R H = 5.294654 × 10 − 11   m (79)

which should represent the electron ground state orbit if the electron should circle, together with its charge, around r₀: indeed the ground-state effective orbit of the electron barycentre observed from m_p is r ′ 0 . Then, writing the (59) as r B = r 0 / ( 1 + ε m ) we get

r B = r 0 1 + m e m P = α / 4 π R H ( 1 + ε m ) = α 4 π R ∞ = 5.291772 × 10 − 11   m (80)

corresponding to the Bohr radius which, on our results, is the orbit of the electron charge referred to B(centre of mass of the system electron-proton)see Figure 9(a).

Now, still referring to conf. (c), the effective electron orbit (i.e. its barycenter) is r ′ 0 where its effective speed is

υ ′ 0 = υ 0 r ′ 0 r 0 = υ 0 ( r 0 + r e ) r B ( 1 + ε m ) = υ 0 [ r B ( 1 + ε m ) + ε r r B ] / r B ( 1 + ε m ) = υ 0 ( 1 + ε m + ε r ) 1 + ε m = α c ( 1 + ε r 1 + ε m ) (81)

and since the distance between the electron charge and the proton charge is r₀ (either considering the electron circling along r ′ 0 or r₀), it turns out that along r ′ 0 the photon admitted frequency is ν ′ 0 = ν 0 , and therefore we get

δ ′ 1 = 2 ν 0 ν ′ e0 = 2 ν 0 / ( υ ′ 0 / 2 π r ′ 0 ) = 2 ν 0 / υ 0 r ′ 0 r 0 / 2 π r ′ 0 = 2 ν 0 2 π r 0 / υ 0 = 1 (82)

meaning one photon every two electron ground-state orbit, and since r ′ 0 = ( r 0 + r e ) is also a circular orbit, the effective electron orbit r ′ n has following parameters:

r ′ n = r ′ 0 n 2 ;     υ ′ n = υ ′ 0 / n ;     ν ′ e n = ν ′ e 0 / n 3 ;     ν ′ n = ν n (83)

thus for each circular orbit we get

δ ′ n = 2 ν n ν ′ e n = ( 2 ν 0 / n 2 ) / ( ν ′ e 0 / n 3 ) = 2 n ν 0 / ν ′ e 0 = n (84)

the same as along r n .

To apply properly the Conservation of momentum (CoM) to the impact photon-electron, the atom nucleus has to be considered fixed as origin in the common centre of gravity nucleus-electrons. On H atom, referring to Figure 9(c), an inelastic impact photon-electron, since photon and w, see Figure 8, have same direction, the (52), gives

w = 2 m c / m r = 2 γ ν c / m r = 2 γ ν c 2 / c m r = 2 h ν / c m r (85)

where the frequency ν, see (70), depends on the effective distance r between the proton and the electron charge, while the term w, see also Figure 10, is the electron radial speed originated by the impact of one photon during the impact time T (=1/ν); for n photons of the same ray the (85) gives

w n = n w = 2 n h ν / c m r (86)

while the generic condition

w = υ (ionization condition) (87)

where υ is the electron speed along the ionization orbit, represents the sufficient condition for an electron to escape from its atom, (with zero final velocity of the electron, like on Absorption effect and, of course, for H atom).

Now, writing the (66) as ν 0 = m r υ 0 2 / 2 h , the ( 85) leads to

w 0 = 2 h m r υ 0 2 2 h c m r = υ 0 2 / c = α 2 c = 15964.35   m / s ≅ c / 137 2 (88)

thus the ionization condition along r₀, for one photon only, becomes

w 0 υ 0 = υ 0 c = α ≅ 1 / 137 ≪ 1 (89)

hence the ionization, requiring w / υ = 1 , could happen for n = 137, but the number of admitted photons, along r₀, is n = 1. Now, along the progressive orbit # n, the admitted photons frequency is ν n = ν 0 / n 2 , and since n is the number of admitted photons along each electron circular d-orbit, we may infer that each progressive orbit is composed of n electron d-orbits, yielding n² admitted photons along each of these orbits.

Referring now to the electron effective ground-state orbit r ′ 0 , regarding the orbit r ′ n ≡ r ′ 0 n 2 , where υ ′ n = υ ′ / n , the (86), for n² photons, gives

w n 2 ( = n w # n ) = n 2 2 h ν ′ n c m r = n 2 2 h ν ′ 0 c m r n 2 = 2 h ν ′ 0 c m r (90)

constant along every circular orbit, and where ν ′ 0 from (66) becomes

ν ′ 0 = m r υ ′ 0 2 / 2 h (91)

which represents the photon virtual frequency, intending the admitted frequency if the electron charge would be coincident with the electron barycenter along its effective orbit r ′ 0 ; indeed, the photons admitted frequency is related to the effective distance r₀, where the effective photon frequency is therefore ν 0 . In fact, we introduce ν ′ 0 to determine the value of υ ′ 0 (ground-state electron effective orbital speed).

Therefore the ionization condition along r ′ n becomes

w n 2 = υ ′ n = υ ′ 0 / n (92)

yielding

2 h ν ′ 0 / c m r = υ ′ 0 / n ⇒ n = υ ′ 0 c m r / 2 h ν ′ 0 (93)

then

n = υ ′ 0 c m r 2 h / 2 h m r υ ′ 0 2 ⇒ n = c υ ′ 0 ≅ c υ 0 = c α c ≅ 137 (94)

but n has to be in integer, so we can infer n = c / υ ′ 0 = 137 and therefore we get

υ ′ 0 = c / 137 (95)

representing the electron speed along its effective ground-state orbit, while

υ ′ 137 = υ ′ 0 / 137 = c / 137 2 (96)

represents the electron speed along its ionization orbit, see also Table 2.

Then, since υ ′ 0 = α c ( 1 + ε r 1 + ε m ) we obtain

c 137 = α c ( 1 + ε m + ε r ) 1 + ε m (97)

then

1 + ε m + ε r = ( 1 + ε m ) 137 α ⇒ ε r = ( 1 + ε m ) ( 1 137 α − 1 ) = 0.0002629141 (98)

and since, for definition, r e = ε r r B , we get

r e = ε r r B = ε r α 4 π R ∞ = ( 1 + ε m ) ( 1 137 α − 1 ) α / 4 π R ∞ = ( 1 137 α − 1 ) α / 4 π R H (99)

and since 4 π R H = α / r 0 we also get

r e = ( 1 137 α − 1 ) r 0 = 1.391281 × 10 − 14 m (electron radius) (100)

Referring to Figure 11 representing the Absorption of photons into a circling electron, let us assume the nucleus mass m_N >> m_e so to consider the nucleus fixed in the atom centre of gravity B.

Now, the expression of the total energy of the system photon-electron is given by

T = E + U r + K e + K r (101)

where E (=mc²) is the energy of the incident light, U_r (=−e²/4πε₀r) is the potential due to electrostatic attraction acting on the electron, K_e (=½m_eυ²) is the electron orbital kinetic energy, and K_r (=½m_ew²) its radial kinetic energy related to its radial speed w due to the impact photon-electron.

Regarding the Absorption/Emission effect (elements on gaseous state), see Figure 11, along circular orbits it is K_r = 0 and moreover, at the end of absorption, along the orbit r₂, (where the photons have been absorbed), it is E₂ = 0; hence between two circular orbits r₁ and r₂, the (101) gives

E 1 + U r 1 + K e 1 = U r 2 + K e 2 . (102)

Now, from (69) U_r = −m_eυ² and since K_e = ½m_eυ², we get ( U r + K e ) = − ½ m e υ 2 thus from (102) we get E 1 − ½ m e υ 1 2 = − ½ m e υ 2 2 ; so, as E₁ = hn, and since m_eυ² = e²/4πε₀r we find

h ν = ½ m e υ 1 2 − ½ m e υ 2 2 = e 2 8 π ε 0 ( 1 r 1 − 1 r 2 ) (103)

Then, according to (71) we have r₁ = r₀n² and r₂ = r₀k² (with k > n as r₂ > r₁), thus

ν = ( e 2 8 π ε 0 h r 0 ) ( 1 n 2 − 1 k 2 ) (104)

and plugging the (74) written as ν₀ = e²/8πε₀hr₀, we find

ν = ν 0 ( 1 n 2 − 1 k 2 ) (105)

which is the photons frequency between two circular orbits, where n represents the progressive specific number of each circular orbit and where k turns out to have, for each of them, the values k = n + 1, n + 2, ∙∙∙, n_i which are, one by one, the number of the remaining external circular orbits, and where n_i is the ionization orbit; on H atom, n_i = 137.

Claimed fall of a circling electron into its nucleus: an electrical current emits an electro-magnetic radiation and therefore it is claimed that the circulating electrical charge of an electron should also emit an e.m. radiation yielding the electron, in a short time, to fall into the nucleus; but on our results, a free electron, moving, for instance, along a copper wire under an electrical potential difference, when entering into an atom influence, (at that moment the electron charge will return to face the atom nucleus), will release the necessary photons to reach the atom energy level corresponding to the energy previously received (during the absorption effect). Indeed, along circular orbits, it is w = 0, therefore the absorption/emission of photons cannot happen along circular orbit, that is why the circling electrons are absorbing/emitting photons only between circular orbits, and therefore, the e.m. radiation is due to the emitted photons during their re-entry toward inner orbits; at this regard, the photons emission is necessary for the electrons not to fall into their nucleus.

Between the electron ground-state orbit r₀ and its extraction orbit r ® ¥ (intending on microscopic scale), the (101), valid for every interaction light-matter, gives

E + U r 0 + K e 0 + K r 0 = E ′ + U r → ∞ + K e ∞ + K ae (106)

where E' is the energy of re-emitted light, K ae = ½ m e w ae 2 the electron kinetic energy after its extraction (intending on macroscopic scale), w_ae its related radial speed, while the other terms have been defined referring to (101).

On ground-state, as also shown between Equations (102) and (103), it is

U r 0 + K e 0 = − m e υ 0 2 + ½ m e υ 0 2 = − ½ m e υ 0 2 = − W f (107)

with υ₀ the electron speed along r₀ and W_f the work function (electron extraction work); now, at the start of impact it is w = 0 giving K r 0 ( = ½ m e w 2 ) = 0 while for r ® ¥, the electron orbital speed υ_¥ ® 0,so

( U r → ∞ + K e 0 ) = − ½ m e υ ∞ 2 → 0 (108)

hence the (106) gives

E − W f = E ′ + K ae ⇒ E = E ′ + W f + K ae (109)

where W f + K ae is the total kinetic energy transferred from light to electron, that is the sum of the sufficient energy to remove the electron plus the one to give it the energy after extraction.

On Photoelectric Effect (PhE), the light scatters off an electron (K_ae ≥ 0), but it is not re-emitted, hence E' = 0, so the (109), with n_f (=W_f/h) the specific threshold frequency, becomes

E = W f + K ae ⇒ h ν = h ν f + ½ m e w ae 2 (110)

showing that for ν = ν_f there is ionization with w_ae = 0.

At frequency ν_f the electron radial speed w_f due to the impact of one photon, see Equation (85), is w f = 2 h ν f / m e c = 2 W f / m e c , and writing the (66) as

υ 0 = ( 2 h ν f / m e ) 1 / 2 = ( 2 W f / m e ) 1 / 2 (111)

we get

w f υ 0 = ( 2 W f m e c ) ( 2 W f m e ) 1 2 = ( 2 W f m e ) ½ / c = υ 0 c ⇒ w f = υ 0 2 c (112)

which is equal to (88), and since the values of W_f are in the range @ 2 - 6 eV, the Equation (112) gives w_f/υ₀ @ 0.0028 - 0.0048 meaning that the ionization (which for elements on solid state where electrons are circling along fixed orbits requires w = υ₀), at frequency ν_f needs n_f photons: indeed, the electron radial speed due to n_f photons with frequency ν_f, see (86), is

w n f = n f w f = n f 2 W f / m e c (113)

so the ionization condition becomes w n f = υ 0 leading to n f 2 W f / m e c = ( 2 W f / m e ) ½ giving

n f = c / 2 W f / m e = c m e / 2 W f = υ 0 / w f = c / υ 0 (114)

that is, on PhE, the number of photons at frequency ν_f necessary for ionization (w_ae = 0). Now, if n_f photons, at frequency ν_f, are sufficient for ionization, then the frequency

n f ν f ≡ ν 1 (115)

is sufficient for ionization (w_ae = 0) with one photon only, so ν₁ is the threshold between PhE and Compton effect which requires one photon only, as shown on next chapter.

To find the number n₁ related to the frequency ν₁ let us equate the (66) written as υ 0 = 2 h ν 1 / m e , to the electron radial speed due to n₁ photons, w n 1 = 2 n h ν 1 / c m e yielding ( 2 h ν 1 / m e ) = ( 2 n 1 h ν 1 / c m e ) 2 and then

n 1 2 = m e c 2 2 h ν 1 = m e c 2 2 h n f ν f = m e c 2 2 n f W f ⇒ n 1 = c m e / 2 n f W f (116)

and plugging W f = c 2 m e / 2 n f 2 as given by (114), we find

n 1 = n f . (117)

For instance as for caesium (Cs), having W_f 1.95 eV, since n f = c ( m e / 2 W f ) ½ ≅ 361.9 we may infer n_f = 361 leading to n₁ = 19, while as for Be or Co, having W_f @ 5 eV we may infer n_f = 225 leading to n₁ = 15.

Here, see Figure 12, the incident photon (length λ, frequency ν), while ejecting a circling electron is also reflected (λ', ν') so the recoiling electron, emitting a photon λ' toward the Observer A, represents a source in motion from A along the direction w, implying an undoubted Doppler effect.

Now, on the basis that the incident photon starts to be reflected at the same time when it starts to hit the electron, and since T ′ = 1 / ν ′ is the emission time of the reflected photon, it turns out that T ′ is also the whole interaction time (and this means that there is not a complete absorption of the incident photon followed by an emission). Now, with T ′ the whole impact time photon-electron, the momentum transferred from the incident light to the electron, as per (53), is

p ′ = m ′ c = γ ν ′ c = γ c / T ′ (118)

and the same value is then transferred from the reflected photon to the electron, so the Conservation of Momentum (CoM) along the direction normal to w, becomes

γ c / T ′ sin θ = γ c / T sin θ ′ (119)

giving θ = θ ′ . Then, the length of the reflected photon, for the Observer A, see (17), is

λ ′ = λ + Δ λ (120)

where Δλ = w_AT', where w_A = wcosθ is the component of the electron speed along the direction of the Observer A, and T ′ = 1 / ν ′ is, for A, the photon transit time, so we get

λ ′ − λ ≡ Δ λ = w T ′ cos θ (121)

Now the CoM along w, being θ = θ ′ , gives

γ c cos θ T ′ + γ c cos θ T ′ = m e w ⇒ w T ′ = ( 2 γ c cos θ ) / m e (122)

so the (121) becomes

λ ′ − λ ( ≡ Δ λ ) = 2 γ c cos 2 θ m e = 2 h cos 2 θ c m e (123)

Now, 2 θ + φ = π , ⇒ θ = ( π − φ ) / 2 , hence cos θ = sin φ / 2 and therefore

Δ λ = 2 h sin 2 ( φ 2 ) / c m e (124)

and since 2 sin 2 ( φ 2 ) = ( 1 − cos φ ) ,we get the Compton equation:

Δ λ = λ ′ − λ = h ( 1 − cos φ ) / c m e (125)

which cannot be obtained via the Doppler effect relativistic equations regarding the light.

We point out that the (122) written as w = ( 2 γ ν ′ c cos θ ) / m e , for cosθ = 1 equals the (85), implying the impact of one photon only.

Now, the (123), for cosθ = 1 gives λ ′ = c ν + 2 γ c m e we can write as:

T ′ = 1 ν + 2 γ m e = ( m e + 2 γ ν ) / ν m e (126)

which plugged into (122) gives

w = ( 2 γ c m e ) / ( m e + 2 γ ν ) / ν m e = 2 γ ν c 2 γ ν + m e = c 1 + m e / 2 γ ν (127)

yielding w → c for ν → ∞.