Jump to ContentJump to Main Navigation
Show Summary Details
More options …

Open Physics

formerly Central European Journal of Physics

Editor-in-Chief: Seidel, Sally

Managing Editor: Lesna-Szreter, Paulina

IMPACT FACTOR 2018: 1.005

CiteScore 2018: 1.01

SCImago Journal Rank (SJR) 2018: 0.237
Source Normalized Impact per Paper (SNIP) 2018: 0.541

ICV 2017: 162.45

Open Access
See all formats and pricing
More options …
Volume 16, Issue 1


Volume 13 (2015)

An attempt to geometrize electromagnetism

XiuLin Huang
  • Corresponding author
  • Changchun Observatory, National Astronomical Observatories, Chinese Academy of Sciences, Changchun, China
  • Email
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Yan Xu / ChengZhi Liu
Published Online: 2018-12-31 | DOI: https://doi.org/10.1515/phys-2018-0106


This study investigates the curved worldline of a charged particle accelerated by an electromagnetic field in flat spacetime. A new metric, which dependes on the charge-to-mass ratio and electromagnetic potential, is proposed to describe the curve characteristic of the world-line. The main result of this paper is that an equivalent equation of the Lorentz equation of motion is put forward based on a 4-dimensional Riemannian manifold defined by the metric. Using the Ricci rotation coefficients, the equivalent equation is self-consistently constructed. Additionally, the Lorentz equation of motion in the non-inertial reference frames is studied with the local Lorentz covariance of the equivalent equation. The model attempts to geometrize classical electromagnetism in the absence of the other interactions, and it is conducive to the establishment of the unified theory between electromagnetism and gravitation.

Keywords: Electromagnetism; Worldline; Riemannian Metric; Ricci rotation coefficients; Local Lorentz transformations

PACS: 03.50.De; 04.20.Cv; 02.40.Ky

1 Introduction

Since Einstein’s general relativity succeeded in geometrizing gravity, numerous people have made great efforts to geometrize electromagnetism in lots of unified-field theories [1, 2, 3, 4, 5, 6, 7, 8]. However, these unified-field theories are not widely accepted. In recent years, work on geometrization of electromagnetism without consideration of gravity has begun to increase [9, 10, 11]. Such studies are not complicated and the physical meaning is more straightforward. Moreover, the research on only geometrization of electromagnetism constantly contributes to establishing a final unified field theory.

Gravity is identified as a geometric phenomenon of a curved spacetime and the geometric phenomenon is the subject investigated in general relativity. Electromagnetism is essentially different from gravity, because it does not relate to the geometric phenomenon of the curved spacetime. Finding a suitable geometric phenomenon for electromagnetism is the first problem for geometrizing electromagnetism. In the previous references [12, 13, 14, 15], the geometrical description of the worldlines of charged particles in homogeneous electromagnetic fields have been studied using Frenet-Serret formulae. The time-like world-line of a charged particle accelerated by an electromagnetic field is possibly a geometric phenomenon of geometrizing electromagnetism. When a particle is accelerated by an electromagnetic force in flat spacetime, the worldline of the particle becomes a curved line xa(τ), and it is closely relate to the Lorentz equation of motion


The geometrical characteristic of xa(τ) may be the key to geometrize electromagneticsm. In this paper, we would like to provide an investigation on the geometrical characteristic of the worldline xa(τ) from a different point of view.

The essential difference between the curved spacetime and the curved worldline indicates that there could be many curved worldlines at one spacetime point. To show the feature of the curved worldline, we structure “artificial” curvilinear coordinates {xâ} regarding the different metrics of the curved worldline xa(τ). The coordinates {xâ} are not real curved spacetime, but are hypothetical curved coordinates associated with the flat coordinates {xa}. Therefore, a metric could be obtained using the linear relationship between the flat differential coordinates dxa and the “artificial" differential coordinates dxâ ,which differs from the model that a metric is derived by the coordinates transformation [16, 17, 18]. In this current work, we attempt to introduce differential geometry by defining the new metric of the curved worldlines of a charge particle in an electromagnetic field, and then obtain an equivalent equation of the Lorentz equation of motion using differential geometry.

This paper is organized as follows. In section 2, a new metric is given depending on the worldline of a charged particle in an electromagnetic field. In section 3, an equivalent equation of the Lorentz equation of motion is proposed using the new metric. section 4 discusses the local Lorentz covariance of the Lorentz equation of motion. section 5 provides a conclusion.

2 The new metric for classical electromagnetism

The length element squared ds2 of a worldline xa(τ) in 4-dimension flat spacetime is


where metric ηab is


The differential coordinate dxa describes the motion of a charged particle in an electromagnetic field and 4-velocity of the charged particle is Ua=(11v2,vi1v2)where vi=dxidx0,i=1,2,3and the speed of light c = 1. Because any curved worldline obeys the relation (2) in flat spacetime, the flat metric ηab does not reflect the curved properties of a worldline. In this model, using “artificial” curvilinear coordinates xâ we give a new metric for the worldline of the charged particle in an electromagnetic field and the effects of radiation are ignored.

Assuming that the transformation between differential coordinates dxâ and dxa is linear, the linear relationship at each point of the worldline is formulated as


Equations (2) and (4) show that the length element ds2 is expressed with the differential coordinate dxâ,




The key issue of obtaining metric Gâb̂ is the relationship between differential coordinates dxâ and dxa.

We consider that the linear transformation haâ needs to satisfy three limiting conditions. In the first limiting condition, in the absence of the case of electromagnetic interaction, the linear transformation haâ should be δa^aThe second limiting condition is provided by the study of the Lorentz equation of motion (1), form which the worldline xa(τ) depends on the charge-to-mass ratio e /m and electromagnetic potential Aa. As a result, the linear transformation haâ should be dependent on e /m and Aa. The third limiting condition derives from the length element’s magnitude change of the worldline, measured by the multiple static-observers positioned at each point of the worldline. The length element of the observers is


which is constant at each point of spacetime. comparing equations (2) and (7), one can find that:


which indicates that the length element’s magnitude is proportional to 1v2. Thus, the linear transformation haâ should show the characteristic of the length element’s magnitude, which is the third limiting condition.

In our model, the linear transformation haâ is defined as:


where Q=Q/(1emA0+Q)and Bâ is expressed as


where Q(t) is a function of time, Aa(xa) is electromagnetic potential and it is used to structure the equation (see Appendix)


Then, the inverse transformation haâ of the transformation haâ yields


The transformation haâ and the inverse transformation haâ satisfy the orthogonal relationship


It is easily demonstrated that the transformation haâ meets the first and second limiting conditions. Besides, using equation (11), we have


which reflects the change of the length element’s magnitude, thus the third limiting condition is satisfied. The new metric Gâ of a worldline’s length element is obtained by equation (6) and (9):


where D0^=h00^=1+emB0^Q,Ba^=Aa1emA0+Q,Qis a function of time and Aa is the electromagnetic potential.

It can be seen that the length element ds2 of the curved worldline xa(τ) has two different expressions with different characteristics: ds2=ηabdxadxbandds2=Ga^b^dxa^dxb^. In the first, ds2 = ηabdxadxb, flat metric ηab is invariant, which does not reflect the curvature of the worldline. The change of length element ds2 at each point along the worldline depends on the differential coordinate dxa, which can describe the motion of the particle. In the second expression, ds2=Ga^b^dxa^dxb^, the “artificial” differential coordinate dsâ is regarded as invariant and it does not describe the motion of the particle. Besides, the change of length element ds2 depends on the metric Gâ at each point of the worldline.

3 Differential geometry for electromagnetism

In this section, we use the length element expression ds2=Ga^b^dxa^dxb^, to derive equations for the motion of a charged particle in an electromagnetic field based on differential geometry. Let M be a 4-dimensional Riemannian manifold with metric Gâ, which is defined by equation (15) at the worldline of the charged particle and Gâ (Q = 0) at other space-time points. Now, the transformation ha â is called tetrad or vierbein. We can derive the equation of a world-line from an action given in the following equation, which is proportional to the length of the worldline.


where proper time τ is a parameter associated with a point on the worldline and the coordinate xâ is variable. Variation of the action (16) with respect to xâ provides the traditional geodesic equation


where Γa^b^c^=12Ga^d^(c^Gd^b^+b^Gc^d^d^Gb^c^)is Christoffel symbol on M. Equation (17) cannot describe the motion of the charged particle, because dxâ/ is not the 4-velocity of the charged particle. We know that dxa/ is considered to describe the motion of a charged particle, and the relationship between dxa and dxâ is given by linear transformation haâ in equation (4). Thus, according to the coordinates xa, equation (17) is rewritten as [19]


where ωabc represents the Ricci rotation coefficients of the tetrad haâ. Besides, the tetrad haâ is also obtained using the metric Gâ under “resting constraint conditions” [20]. In the torsion-free case, ωabc is given as reported in literature [21]




where equation c(haa^hab^)=c(δa^b^)=0anda=hab^b^are used. Using the tetrad (9) and equaton (14), equaton (18) becomes


where F ab = aAbbAa and Aa is electromagnetic potential. Therefore, the Lorentz equation of motion is obtained using 4-dimensional Riemannian manifold M defined by the metric Gâ.

4 Covariance of the Lorentz equation of motion

Motion equation (18) is form invariant with respect to local Lorentz transformations, under which [22], it is expressed as


where ωa' b'c' transform as [21]


where Λa' a are local Lorentz transformations, and the relationship between dxa' and dxâ is given by the Λa' a,


Equation (22) is considered to describe the equation of motion in the different reference frame with coordinates xa. Thus, because equation (18) is equivalent to the Lorentz equation of motion, the Lorentz equation of motion in the other reference frames can be studied using equation (22) and two cases are as follows.

For the case that the local Lorentz transformations are constant, equation (22) becomes


where Fab=ΛaaΛbbFabequation (25) is the Lorenz equation of motion in another inertial frame. When the local Lorentz transformations are time-dependent, the reference frame with coordinates xa' is a non-inertial frame. It is of significant importance that the motion equation in the non-inertial frame is studied using equation (22) depending on the corresponding local Lorentz transformations.

We consider a special non-inertial frame—the proper reference frame (comoving frame) that is attached to the charged particle in an electromagnetic field. Besides, the charged particle in the proper reference frame is always stationary. Thus, dxa' is the differential coordinates of the proper reference frame, and dx0' = and dxi' = 0 in the proper reference frame. From the relation Λaadxa=dxawe have which becomes


The term ΛaacΛbadxbdτdxcdτin equation (22) is calculated,


In the proper frame, equation (22) becomes


Using the Lorentz equation of motion, the latter two items in equation (28) satisfy equation


Thus, motion equation (22) becomes


We know that the proper reference frame (comoving frame) is the non-inertial reference frame, in which the charged particle is subjected to a fictitious force and a electromagnetic force. These two forces cancel each other out which result in the motion equation of the particle as d2xadτ2=0in the proper reference frame. In our method, the motion equation of the particle is expressed as motion equation (22) in the proper reference frame. After being calculated, motion equation (22) becomes motion equation (28), in which the term Λaad2xa/dτ2is fictitious acceleration and the term emFabdxb/dτis the electromagnetic acceleration, and these two forces cancel each other out. The result is reasonable in the proper reference frame.

5 Conclusions

To conclude, this work proposes the new metric (15) in flat spacetime to use for the curved worldline xa(τ) of charged particles in an electromagnetic field. Then, we show that equation (18) is the motion equation of the charged particles and the equation is equivalent to the Lorentz equation of motion using the differential geometry of the metric (15). As the Lorentz equation of motion can describe the change of worldline xa(τ), the motion equation (18) obtained shows that our model is self-consistent. Additionally,we have also investigated the local Lorentz covariance of the Lorentz equation of motion based on the equivalent equation (18), which is form invariant under the local Lorentz transformations. The important conclusion is that the Lorentz equation of motion in different inertial or non-inertial reference frames is related using equation (22). One example is that the motion equation (22) in the proper frame has been discussed and a rational prediction result has been obtained. In a sense, our method extends the Lorentz covariance of the Lorentz equation of motion to the local Lorentz covariance. Admittedly, local Lorentz covariance closely relates to general covariance in general relativity. Despite that general covariance has been proposed for numerous years, physicists do not have a unified understanding and the discussion continues for general covariance [23, 24, 25, 26, 27, 28]. This research may provide a new point of view for understanding geometrizing electromagnetism and the relationship between electromagnetic interaction and general covariance.

The innovation of this paper is that the curve characteristic of the worldline of charged particles in an electromagnetic field is described by the metric (15), and the Lorentz equation of motion is achieved by using the Riemannian geometry defined by the metric. Our method has an interesting relationship with general relativity. The important similarity between general relativity and our method is that a 4-dimensional Riemannian manifold is used for studying interactions. Our method significantly differs from general relativity in the following ways:

  1. Our method investigates the relative motion of an object in flat spacetime. We introduce differential geometry to describe the curved worldline rather than curved spacetime. In the present method, the coordinates {xâ} represent the characteristics of a “curved” coordinate with respect to flat coordinates {xa}, but the coordinates {xâ} are hypothetical.

  2. In general relativity, the metric does not depend on the properties of the object that is accelerated by a gravity indicating that objects of different masses in the gravitational effect have the same geodesic equation (equation of motion). However, in our method, the metric depends on the particle’s charge-to-mass ratio e/m properties. These results are significant, as charged particles with different e /m ratios in an electromagnetic field have different worldlines, consequently introducing different metrics and different motion equations.

  3. In our method, the motion equation (18) is constructed based on the Ricci rotation coefficients rather than the Christoffel symbols. The Ricci rotation coefficients are employed to describe the physical laws in flat spacetime.

To some extent, the present research is significant to unify electromagnetism and gravitation with differential geometry.

Further studies of differential geometry are needed to perfect our model. At first, the study of the classical dynamics equation may be extended into quantum electrodynamics using differential geometry. Secondly, the problem that electromagnetic potential satisfies the Maxwell equation need to be studied in the framework of differential geometry. We believe that the methods and proofs described in this paper bring us a step closer to the final unified theory which will tightly associate differential geometry with fundamental electromagnetic interactions.


The work is supported by Open Foundation of Guizhou Provincial Key Laboratory of Radio Astronomy and Data Processing, Youth Innovation Promotion Association, Chinese Academy of Sciences under Grant No. 2016056, the Development Project of Science and Technology of Jilin Province under Grant No. 20180520077JH and the National Natural Science Foundation of China Nos. 11805022 and 11803057.

Appendix A

1v2is associated with the electromagnetic potential of a charged particle accelerated by an electromagnetic field. According to relativistic mechanics [29],




where U0=1/1v2and Ui=vi/1v2Combining equation (A.1) and (A.2), it can see that U0 satisfies the relationship


U0 can be divided into the contributions of scalar field A0 and vector field Ai, that is U0=U0e+U0m,dU0e=emA0xidxiand dU0m=emAix0dxiFor the interaction of a charged particle with a scalar field A0, the following relationship is obtained,


where C is a constant determined by the initial conditions. However, U0m is not easy to obtain. It is considered that the effect of vector field Ai in equation (A.3) can always be replaced by a a function of time at some fixed spacetime points. Therefore, U0m is rewritten as a function of time. We set


where Q(t) is the time function and determined by the vector field Ai. Based on the above argument, 1v2becomes a function of the electric potential A0 and time t.


  • [1]

    Goenner H.F., On the history of unified field theories, Living Reviews in Relativity, 2004, 7, 1830-1923. Web of ScienceGoogle Scholar

  • [2]

    Witten L., Geometry of gravitation and electromagnetism, Phys. Rev., 1959, 115, 206-214. CrossrefGoogle Scholar

  • [3]

    Borchsenius K., Unified theory of gravitation, electromagnetism, and the yang-mills field, Phys. Rev. D, 1976, 13, 2707-2712. CrossrefGoogle Scholar

  • [4]

    Ferraris M., Kijowski J., Unified geometric theory of electromagnetic and gravitational interactions, General Relat. Gravit., 1982, 14, 37-47. Google Scholar

  • [5]

    Vargas J. G., On the geometrization of electrodynamics, Found. Phys., 1991, 21, 379-401. Google Scholar

  • [6]

    Popławski N.J., Geometrization of electromagnetism in tetrad-spin gravity, Modern Phys. Let. A, 2009, 24, 431-442. Google Scholar

  • [7]

    Popławski N.J., Torsion as electromagnetism and spin, International J. Theor. Phys., 2010, 49, 1481-1488. CrossrefGoogle Scholar

  • [8]

    Giglio J.F.T., Rodrigues Jr W.A., Gravitation and electromagnetism as geometrical objects of a riemann-cartan spacetime structure, Adv. Appl. Clifford Algebras, 2012, 22, 649-664. CrossrefGoogle Scholar

  • [9]

    Shahverdiyev S.S., General geometry and geometry of electromagnetism, 2002, arXiv preprint hep-th/0205224. Google Scholar

  • [10]

    Olkhov O.A., New approach to geometrization of electromagnetic field, Amer. J. Phys. Appl., 2015, 3, 221-225. Google Scholar

  • [11]

    Duarte C.A., The classical geometrization of the electromagnetism, Int. J. Geom. Meth. Modern Phys., 2015, 12, 1560022. Google Scholar

  • [12]

    Synge J.L., Timelike helices in flat space-time, Proc. Royal Irish Acad., 1966, 65, 27-42. Google Scholar

  • [13]

    Honig E., Schucking E.L., Vishveshwara C.V., Motion of charged particles in homogeneous electromagnetic fields, J. Math. Phys., 1974, 15, 774-781. CrossrefGoogle Scholar

  • [14]

    Caltenco J.H., Linares y M., Lopez-Bonilla J.L., Intrinsic geometry of curves and the lorentz equation, Czechoslovak J. Phys., 2002, 52, 839-842. CrossrefGoogle Scholar

  • [15]

    Formiga J.B., Romero C., On the differential geometry of time-like curves in minkowski spacetime, Amer. J. Phys., 2006, 74, 1012-1016. Google Scholar

  • [16]

    Misner C.W., Thorne K.S., Wheeler, J.A., 1973, San Francisco, W.H. Freeman and Co. 

  • [17]

    Ni W.T., Zimmermann M., Inertial and gravitational effects in the proper reference frame of an accelerated, rotating observer, Phys. Rev. D, 1978, 17, 1473-1476. Google Scholar

  • [18]

    Nelson R.A., Generalized lorentz transformation for an accel- erated, rotating frame of reference, J. Math. Phys., 1987, 28, 2379-2383. CrossrefGoogle Scholar

  • [19]

    Aldrovandi R., Barros P., Pereira J., The equivalence principle revisited, 2003, Found. Phys., 33, 545-575. CrossrefGoogle Scholar

  • [20]

    Mei T., A new variable in general relativity and its applications for classic and quantum gravity, 2006, arXiv preprint grqc/0611063. 

  • [21]

    De Felice F., Clarke C.J., Relativity on curved manifolds, Cambridge University Press, 1992. Google Scholar

  • [22]

    Carroll S.M., Lecture notes on general relativity, 1997, arXiv preprint gr-qc/9712019. Google Scholar

  • [23]

    Ellis G.F.R., Matravers D.R., General covariance in general relativity? General Relat. Gravit., 1995, 27, 777-788. CrossrefGoogle Scholar

  • [24]

    Zalaletdinov R., Tavakol R., Ellis G.F.R., On general and restricted covariance in general relativity, General Relat. Gravit., 1996, 28, 1251-1267. Google Scholar

  • [25]

    Gron O., Voyenli K., On the foundation of the principle of relativity, Foundations of Physics, 1999, 29, 1695-1733. CrossrefGoogle Scholar

  • [26]

    Brown H.R., Physical relativity: Space-time structure from a dynamical perspective, 2005, Oxford University Press on Demand. Google Scholar

  • [27]

    Westman H., Sonego S., Events and observables in generally invariant spacetime theories, Found. Phys., 2008, 38, 908-915. CrossrefWeb of ScienceGoogle Scholar

  • [28]

    Chamorro A., On the meaning of the principle of general covariance, Int. J. Theor. Phys., 2013, 52, 117-129. CrossrefGoogle Scholar

  • [29]

    Pauli W., Theory of relativity, 1958, Courier Dover Publications. Google Scholar

About the article

Guizhou Provincial Key Laboratory of Radio Astronomy and Data Processing;

Received: 2018-05-16

Accepted: 2018-09-06

Published Online: 2018-12-31

Citation Information: Open Physics, Volume 16, Issue 1, Pages 848–853, ISSN (Online) 2391-5471, DOI: https://doi.org/10.1515/phys-2018-0106.

Export Citation

© 2018 X. Huang et al., published by De Gruyter. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. BY-NC-ND 4.0

Comments (0)

Please log in or register to comment.
Log in