The textbooks say nothing can travel faster than light, not even light itself. New experiments show that this is no longer true, raising questions about the maximum speed at which we can send information.
Can a light pulse travel faster than the speed of light? This question has intrigued physicists for many years because such an event could violate Einstein's theory of special relativity and the principle of causality (that 'cause' always precedes 'effect'). Together these imply that no object or information can travel faster than the speed of light, c=3times108 m s-1. For nearly two decades, physicists have been sending certain light pulses faster than c over short distances (so-called superluminal propagation), but the light pulses have always been distorted in the process so interpreting these experiments has been difficult1-3.
In May this year, Mugnai et al.4 reported superluminal behaviour in the propagation of microwaves (centimetre wavelengths) over much longer distances (tens of centimetres) at a speed 7% faster than c. A report by Wang et al.5 ( page 277 of this issue) now demonstrates a very large superluminal effect for pulses of visible light, in which a pulse propagates in a specially prepared medium with a negative velocity of -c/310: that is, not only faster than a pulse travelling in a vacuum, but so fast that the peak of the pulse exits the medium before it enters it!
A negative velocity can be understood by comparing the times it would take for identical pulses of light to cover some distance L in a vacuum (travelling at velocity c) and in a superluminal medium (travelling at velocity v). The difference in transit times DeltaT= L/v-L/c is a negative quantity if the velocity is superluminal. If v has a negative value then DeltaT can become sufficiently negative that the peak of the pulse emerges from the medium at an instant earlier than when the peak of the pulse enters. This brings to mind Arthur Buller's well-known limerick with relativistic over tones:
There was a young lady named Bright,
Whose speed was far faster than light;
She set out one day,
In a relative way,
And returned home the previous night.
But Wang et al.5 claim that, unlike the heroine of this rhyme, their light pulses do not violate causality. They argue that their superluminal pulses are the result of the wave nature of light itself (fortunately, making it impossible for an object with mass to travel faster than c) and that no actual information, or signal, is transmitted faster than c. They use smooth, well-defined light pulses, so that the peak of the pulse at the output results from the forward rising edge of the input pulse, which occurs far earlier in time, making it consistent with causality. An abrupt feature in the light pulse would not be able to travel faster than c. This means that even if the 'effect' appears to precede the 'cause', you still can't send useful information such as news of an impending accident faster than c.
A light pulse has a finite duration, and it is a well-known theorem in physics (the bandwidth theorem) that, to create a pulse of finite duration, an infinite number of waves of different frequency must be added together. The shorter the desired pulse, the larger the bandwidth of frequencies that must be used. All light pulses are therefore formed by a packet of waves of different frequency, each of which has a different amplitude and phase. There is a distinction between the speed of individual waves, called the phase velocity, vp, and the velocity at which the peak of the wavepacket propagates, known as the group velocity, vg. In a vacuum the phase and group velocities are the same, but in a highly absorbing or dispersive medium they are usually different. A negative group velocity results when the phases of the different frequency components are shifted by the medium through which they travel, so that the wavepacket they form at the exit is brought forward in time compared with the same pulse travelling through a vacuum.
One way to achieve negative velocity is to modify the refractive index of the medium through which the light passes. Last year scientists at Harvard6 and elsewhere succeeded in modifying the refractive properties of a cloud of ultracold atoms to generate very slow light pulses with group velocities of a few metres per second. To create the opposite effect superluminal pulses of light you need a medium in which the refractive index changes rapidly, for example near an atomic absorption frequency ( Fig. 1a). The only problem is that the so-called anomalous dispersion region in Fig. 1a, where vg can be negative3, is also in a region where there is increased light absorption. In experiments with such highly absorbing materials, the light pulses are either strongly distorted or absorbed, making any faster-than-light claims difficult to interpret.
Figure 1 Sending photons faster than light. Full legend
High resolution image and legend (95k)
A more promising approach to making superluminal light pulses is to work with an atomic medium where there is gain (amplification of light waves) at the atomic transition frequency. This is achieved in a laser-type medium by creating a 'population inversion', whereby a higher population of atoms are in the excited than in the lower-energy atomic state7. In this case, anomalous dispersion occurs at frequencies lower than the transition frequency. But close to the transition frequency, where the effect is largest, the rapidly changing gradient in the refractive index causes severe pulse distortion. One way round this problem is to use a gain doublet8 that is, two closely spaced regions of gain where the zone between has steep anomalous dispersion but without strong pulse distortion (Fig. 1b). This is what Wang et al.5 have now achieved.
The experiment by Wang and co-workers creates this type of gain doublet in a six-centimetre cell containing caesium gas by using two laser fields closely spaced in frequency (see Fig. 1a on page 277). They first measured the refractive index of the caesium using a third 'probe' laser, and produced a dispersion curve similar to Fig. 1b, with a steep gradient in the anomalous dispersion region corresponding to an expected vg= - c/310. When they sent a 3.7-microsecond light pulse through the medium, it appeared at the exit of the cell before it arrived at the entrance. Although the pulse itself is only shifted forward in time by a modest fraction (1.7%) of its width, this corresponds to the wavepacket leaving the cell 62 nanoseconds before it arrives in other words, travelling nearly 20 metres away from the cell before the incoming pulse enters it. Compared with the time to travel six centimetres in a vacuum (about 0.2 nanoseconds), the 62-nanosecond lead means that the group velocity of the pulse inside the medium is - c/310, close to the predicted value.
In this experiment, each of the different frequency components making up the pulse experiences a slightly different dispersion in the medium. The relative phases between them are therefore changed and the pulse shape is shifted to bring the pulse wavepacket (or group velocity) forward in time. So the anomalous dispersion leads to interference between different frequency components of the pulse that produce the superluminal effect. Although amazing, this type of superluminal pulse propagation does not violate the principle of causality.
There remains, however, some debate about what is the true speed at which information is carried by a light pulse. Traditionally the signal velocity of a light pulse is defined as the speed at which the half peak-intensity point on the rising edge of the waveform travels; in this experiment, this is clearly superluminal. In contrast, some researchers argue that the true speed at which information is carried by a light pulse is not the group velocity of a smooth pulse, but rather the speed at which a sudden step-like feature in the waveform travels, which so far has not been shown to exceed c. Superluminal effects are especially interesting in the case of light pulses consisting of only a few photons, in which it could be argued that the group velocity is the same as the velocity of the individual photons. The type of superluminal behaviour discussed here is also predicted to apply to single photons8, which might have implications for the transmission of quantum information.
1. Steinberg, A. M. , Kwiat, P. G. & Chiao, R. Y. Phys. Rev. Lett. 71, 708 711 (1993). | Article | PubMed | ISI | ChemPort |
2. Balcou, Ph. & Dutriax, L. Phys. Rev. Lett. 78, 851854 (1997). | Article | ISI | ChemPort |
3. Chu, S. & Wong, S. Phys. Rev. Lett. 48, 738741 (1982). | Article | ISI | ChemPort |
4. Mugnai, D. , Ranfagni, A. & Ruggeri, R. Phys. Rev. Lett. 84, 4830 4833 (2000). | Article | PubMed | ISI | ChemPort |
5. Wang, L. J. , Kuzmich, A. & Dogariu, A. Nature 406, 277 279 (2000). | Article | PubMed | ISI | ChemPort |
6. Hau, L. V. , Harris, S. E. , Dutton, Z. & Behroozi, C. H. Nature 397, 594598 ( 1999). | Article | ISI | ChemPort |
7. Chiao, R. Y. Phys. Rev. A 48, R34R38 (1993). | Article | PubMed | ISI | ChemPort |
8. Steinberg, A. M. & Chiao, R. Y. Phys. Rev. A 49, 20712075 ( 1994). | Article | PubMed | ISI |
http://www.nature.com/nature/journal...6243a0_F1.html
http://www.nature.com/cgi-taf/DynaPa...6243a0_fs.html
----------------------------
Werent we discussing about universe and travelling faster than light some time?? :)
Can a light pulse travel faster than the speed of light? This question has intrigued physicists for many years because such an event could violate Einstein's theory of special relativity and the principle of causality (that 'cause' always precedes 'effect'). Together these imply that no object or information can travel faster than the speed of light, c=3times108 m s-1. For nearly two decades, physicists have been sending certain light pulses faster than c over short distances (so-called superluminal propagation), but the light pulses have always been distorted in the process so interpreting these experiments has been difficult1-3.
In May this year, Mugnai et al.4 reported superluminal behaviour in the propagation of microwaves (centimetre wavelengths) over much longer distances (tens of centimetres) at a speed 7% faster than c. A report by Wang et al.5 ( page 277 of this issue) now demonstrates a very large superluminal effect for pulses of visible light, in which a pulse propagates in a specially prepared medium with a negative velocity of -c/310: that is, not only faster than a pulse travelling in a vacuum, but so fast that the peak of the pulse exits the medium before it enters it!
A negative velocity can be understood by comparing the times it would take for identical pulses of light to cover some distance L in a vacuum (travelling at velocity c) and in a superluminal medium (travelling at velocity v). The difference in transit times DeltaT= L/v-L/c is a negative quantity if the velocity is superluminal. If v has a negative value then DeltaT can become sufficiently negative that the peak of the pulse emerges from the medium at an instant earlier than when the peak of the pulse enters. This brings to mind Arthur Buller's well-known limerick with relativistic over tones:
There was a young lady named Bright,
Whose speed was far faster than light;
She set out one day,
In a relative way,
And returned home the previous night.
But Wang et al.5 claim that, unlike the heroine of this rhyme, their light pulses do not violate causality. They argue that their superluminal pulses are the result of the wave nature of light itself (fortunately, making it impossible for an object with mass to travel faster than c) and that no actual information, or signal, is transmitted faster than c. They use smooth, well-defined light pulses, so that the peak of the pulse at the output results from the forward rising edge of the input pulse, which occurs far earlier in time, making it consistent with causality. An abrupt feature in the light pulse would not be able to travel faster than c. This means that even if the 'effect' appears to precede the 'cause', you still can't send useful information such as news of an impending accident faster than c.
A light pulse has a finite duration, and it is a well-known theorem in physics (the bandwidth theorem) that, to create a pulse of finite duration, an infinite number of waves of different frequency must be added together. The shorter the desired pulse, the larger the bandwidth of frequencies that must be used. All light pulses are therefore formed by a packet of waves of different frequency, each of which has a different amplitude and phase. There is a distinction between the speed of individual waves, called the phase velocity, vp, and the velocity at which the peak of the wavepacket propagates, known as the group velocity, vg. In a vacuum the phase and group velocities are the same, but in a highly absorbing or dispersive medium they are usually different. A negative group velocity results when the phases of the different frequency components are shifted by the medium through which they travel, so that the wavepacket they form at the exit is brought forward in time compared with the same pulse travelling through a vacuum.
One way to achieve negative velocity is to modify the refractive index of the medium through which the light passes. Last year scientists at Harvard6 and elsewhere succeeded in modifying the refractive properties of a cloud of ultracold atoms to generate very slow light pulses with group velocities of a few metres per second. To create the opposite effect superluminal pulses of light you need a medium in which the refractive index changes rapidly, for example near an atomic absorption frequency ( Fig. 1a). The only problem is that the so-called anomalous dispersion region in Fig. 1a, where vg can be negative3, is also in a region where there is increased light absorption. In experiments with such highly absorbing materials, the light pulses are either strongly distorted or absorbed, making any faster-than-light claims difficult to interpret.
Figure 1 Sending photons faster than light. Full legend
High resolution image and legend (95k)
A more promising approach to making superluminal light pulses is to work with an atomic medium where there is gain (amplification of light waves) at the atomic transition frequency. This is achieved in a laser-type medium by creating a 'population inversion', whereby a higher population of atoms are in the excited than in the lower-energy atomic state7. In this case, anomalous dispersion occurs at frequencies lower than the transition frequency. But close to the transition frequency, where the effect is largest, the rapidly changing gradient in the refractive index causes severe pulse distortion. One way round this problem is to use a gain doublet8 that is, two closely spaced regions of gain where the zone between has steep anomalous dispersion but without strong pulse distortion (Fig. 1b). This is what Wang et al.5 have now achieved.
The experiment by Wang and co-workers creates this type of gain doublet in a six-centimetre cell containing caesium gas by using two laser fields closely spaced in frequency (see Fig. 1a on page 277). They first measured the refractive index of the caesium using a third 'probe' laser, and produced a dispersion curve similar to Fig. 1b, with a steep gradient in the anomalous dispersion region corresponding to an expected vg= - c/310. When they sent a 3.7-microsecond light pulse through the medium, it appeared at the exit of the cell before it arrived at the entrance. Although the pulse itself is only shifted forward in time by a modest fraction (1.7%) of its width, this corresponds to the wavepacket leaving the cell 62 nanoseconds before it arrives in other words, travelling nearly 20 metres away from the cell before the incoming pulse enters it. Compared with the time to travel six centimetres in a vacuum (about 0.2 nanoseconds), the 62-nanosecond lead means that the group velocity of the pulse inside the medium is - c/310, close to the predicted value.
In this experiment, each of the different frequency components making up the pulse experiences a slightly different dispersion in the medium. The relative phases between them are therefore changed and the pulse shape is shifted to bring the pulse wavepacket (or group velocity) forward in time. So the anomalous dispersion leads to interference between different frequency components of the pulse that produce the superluminal effect. Although amazing, this type of superluminal pulse propagation does not violate the principle of causality.
There remains, however, some debate about what is the true speed at which information is carried by a light pulse. Traditionally the signal velocity of a light pulse is defined as the speed at which the half peak-intensity point on the rising edge of the waveform travels; in this experiment, this is clearly superluminal. In contrast, some researchers argue that the true speed at which information is carried by a light pulse is not the group velocity of a smooth pulse, but rather the speed at which a sudden step-like feature in the waveform travels, which so far has not been shown to exceed c. Superluminal effects are especially interesting in the case of light pulses consisting of only a few photons, in which it could be argued that the group velocity is the same as the velocity of the individual photons. The type of superluminal behaviour discussed here is also predicted to apply to single photons8, which might have implications for the transmission of quantum information.
1. Steinberg, A. M. , Kwiat, P. G. & Chiao, R. Y. Phys. Rev. Lett. 71, 708 711 (1993). | Article | PubMed | ISI | ChemPort |
2. Balcou, Ph. & Dutriax, L. Phys. Rev. Lett. 78, 851854 (1997). | Article | ISI | ChemPort |
3. Chu, S. & Wong, S. Phys. Rev. Lett. 48, 738741 (1982). | Article | ISI | ChemPort |
4. Mugnai, D. , Ranfagni, A. & Ruggeri, R. Phys. Rev. Lett. 84, 4830 4833 (2000). | Article | PubMed | ISI | ChemPort |
5. Wang, L. J. , Kuzmich, A. & Dogariu, A. Nature 406, 277 279 (2000). | Article | PubMed | ISI | ChemPort |
6. Hau, L. V. , Harris, S. E. , Dutton, Z. & Behroozi, C. H. Nature 397, 594598 ( 1999). | Article | ISI | ChemPort |
7. Chiao, R. Y. Phys. Rev. A 48, R34R38 (1993). | Article | PubMed | ISI | ChemPort |
8. Steinberg, A. M. & Chiao, R. Y. Phys. Rev. A 49, 20712075 ( 1994). | Article | PubMed | ISI |
http://www.nature.com/nature/journal...6243a0_F1.html
http://www.nature.com/cgi-taf/DynaPa...6243a0_fs.html
----------------------------
Werent we discussing about universe and travelling faster than light some time?? :)
Comment