1 Introduction

Surface plasmon polaritons (SPP) are coupled quasiparticles of photons and charge-density fluctuation, which can be beyond the diffraction limit [1, 2]. This remarkable property provides a way to explore the light–matter interactions at the nanoscale and plays a crucial role in various applications, including super-resolution imaging [2,3,4], nanolithography [5, 6] and the design of highly integrated plasmonic devices [2, 7, 8]. By virtue of the SPP’s nanoscopic and dissipative nature, the spatiotemporal imaging in nano-femto scale is often the optimal route to understand the underlying physical processes of exciting phenomena such as SPP propagation through plasmonic waveguides [9, 10] and verifying the functionalities of the plasmonic device [10,11,12].

More importantly, due to the particular boundary conditions, SPP is a transverse magnetic (TM) wave and has transverse and longitudinal components of the electric field within the plane of propagation [13]. They play significant but different roles in the fields of applied and basic research, respectively. For example, simulation proves that the focusing performance of the planar plasmonic lens and the interference pattern of the pure SPP are dominated by the transverse component of SPP [14,15,16], whereas the longitudinal component of SPP field becomes more pronounced in the SPP superfocusing by a metallic cone [17, 18]. Moreover, the theory previously identified transverse spin angular momentum of SPP arising from the constant instantaneous phase difference of π/2 of the transverse and longitudinal components of SPP electric field, which generates a rotation of the electric-field vector within the propagation plane [13, 19]. Therefore, it is necessary to carry out experiments on the spatiotemporal imaging of the longitudinal and transverse components of the same SPP near-fields in nano-femtosecond scale to further reveal the physical nature of the SPP and to expand its application. Meanwhile, this is also the premise for the experimental mapping of a three-dimensional space–time field of SPP.

In recent years, several potential solutions for measuring the static information of the longitudinal and transverse components of SPP near-fields have been proposed, including the use of the scanning near-field microscope with a custom complex probe [20, 21] and normal incident photoemission electron microscope [22, 23]. Specifically, Meyer zu-Heringdorf and co-workers [22] used normal-incidence light to demonstrate that the photoemission yield produced by SPP–light interference depends on a favorable alignment of the longitudinal field vectors, either parallel or anti-parallel. In their study, by employing a normal incidence scheme, only the interference signal of the longitudinal component could be observed due to eliminating the transverse component of the probe beam.

Based on the superposition principle of the near-field, recently we reported an interferometric time-resolved photoemission electron microscopy (ITR-PEEM) study for SPP from trench structure excited by oblique incident field, which was used to measure the phase difference between the longitudinal and transverse components of SPP at Ag/vacuum interface [24]. Nevertheless, due to the long coherent superposition time of the probe laser pulse with the pump-excited SPP(both pump and probe pulses are with of duration of 130 fs) [24, 25], the interference patterns fill the entire probe area which is difficult to show the intuitively dynamic propagation of SPP near-fields in the same probe area and in the experiment, the position of the probe light and the delay time of the pump-probe pulse need to be adjusted to ensure the capture of dynamic information of the respective components of SPP near-fields. This hinders the capture of the clear spatiotemporal information of the SPP near-field in nano-femto scale. Besides, reference 24 provides only a qualitative explanation for the feasibility of the spatiotemporal imaging of the longitudinal and transverse components of the SPP near-field and it lacks systematic simulation analysis.

In this paper, the spatiotemporal imaging of the few-cycle SPP in nano-Femto scale was performed under two different excitation schemes that include the incident light along and vertical to the long axis of the trench. Experiments demonstrate that the intuitively dynamic propagation of the transverse and longitudinal components evolution of the identical SPP field in nano-Femto scale was captured using the few-cycle (sub-10 fs) pulses for ITR-PEEM experiment of the incident light along the long axis of the trench, which has a short coherent superposition time of the probe laser pulse with the pump-excited SPP. The contributions of the transverse and longitudinal components of SPP near-fields to interference patterns were analyzed by the analytical calculation using the complex dielectric function of silver and FDTD simulation. Meanwhile, it is found that the photoemission intensity induced by the interference of SPP and incident light longitudinal field is far lower than that induced by the interference of the transverse field, and the detectable longitudinal field of SPP is concentrated near the peak of the SPP envelope. Based on Fresnel’s law and the wave theory of the electromagnetic field, the influencing factors of the result are discussed, and the ratio between transverse and longitudinal electric field components is measured for the first time using ITR-PEEM. To the best of our knowledge, we are the first to realize independent spatiotemporal imaging of the transverse and longitudinal components of SPP near-fields in nano-femtosecond scale.

2 Experimental details

The multiphoton photoemission from a superposition of SPP and laser field is recorded using a photoemission electron microscope (Focus GmbH). Typically, the spatial resolution of PEEM, as defined by edge contrast, is better than 40 nm for a simple electrostatic lens column without aberration correction. We used either a mercury lamp or a Ti: sapphire laser excitation source, a repetition rate of 75 MHz, a center wavelength of λ = 800 nm, and a spectral bandwidth from 600 to 900 nm (FEMTOLASERS; Rainbow) as sources of excitation. To implement the spatiotemporal imaging of the longitudinal and transverse components of SPP near-fields in the Femto-nanoscale, the chirp mirror pairs and wedge pairs are used to compensate for dispersion to achieve the shortest possible pulse duration in the PEEM chamber. The second-order autocorrelation curve of the incident laser pulse measured by the autocorrelator (FEMTO LASERS) that includes all the optical components dispersion before reaching the sample is shown in Fig. 1. The pulse duration is about 9 fs. The incident laser is focused onto the sample surface using a 20 cm focal length off-axis parabolic mirror and elliptically shaped focused laser spots are observed featuring major/minor axes of 60/30 μm. Meanwhile, the laser power of 60∼160mW is used in the experiment. For ITR-PEEM experiments, the pulses are interferometrically locked using a Mach–Zender interferometer with each arm having separate polarization control by half-wave plate, and control the position of the probe light pulse by adjusting the beam combiner. The details of the experimental setup have been described in an earlier report [26, 27].

Fig. 1
figure 1

Measured second-order interferometric autocorrelation curve of the incident laser pulse

Full size image

The rectangular 10 × 1 μm2 and 20 × 1 μm2 trench coupling structure are milled into thicknesses ~ 100 nm silver thin film on a clean SiO2 substrate using a focused ion beam (FIB) lithography. The near-field distribution of the trench structure was simulated using a commercial FDTD package (Lumerical FDTD Solutions). The calculations employ a total field scattered field plane wave source. The electric field monitor is placed 5 nm above the silver film. The dielectric permittivity of silver is taken from Johnson and Christy [28].

3 Results and discussion

Figure 2a displays the ITR-PEEM experimental schematic illustration of the incident light perpendicular to the long axis of the trench. To avoid the entanglement of the interference patterns (the static pattern and the dynamic pattern) in the PEEM image, we separated the probe spatially from the pump in the ITR-PEEM schemes, which allows the purely spatiotemporal information of the SPP to be captured by adjusting the time delay (ΔT) between the pump and probe pulses, as shown in Fig. 2a. Throughout the paper, the femtosecond laser pulses illuminate the sample along the y-axis direction at an incidence angle of 65° with respect to the surface normal from the left side which is determined by the PEEM instrument. kL, kS, and kB represent the wave vectors of the laser in-plane, SPP, and interference pattern, respectively.

Fig. 2
figure 2

ITR-PEEM experimental schematic illustration of the incident light perpendicular to the long axis of the trench (a). The PEEM image of the rectangular 10 × 1μm2 trench structure is illuminated by a mercury lamp source (one-photon photoemission) (b). ITR-PEEM images with different delays. P-polarized pump pulse is used efficiently to excite the SPP, and the p- (c) or s-polarized (d) probe pulse interrogates the SPP at a remote point. The time-resolved PEEM images (c) under different time delays were captured using p-polarized probe pulses. The vectors diagram of the laser field and the SPP field is given at the bottom in (c, d). Yellow dashed ovals represent the approximate positions of the pump and probe beams. Zero delays are defined as the time in which SPP of the pump launching and probe pulses of p-polarized are overlapped in the edge of the probe region, as shown in figure (d)

Full size image

Figure 2b shows the PEEM image of the rectangular 10 × 1μm2 trench structure illuminated by a mercury lamp source. The cutoff photon energy of a mercury lamp is 4.9 eV, and typically, the work function of the silver film is about 4.2 eV. The one-photon PEEM image obtained under Hg light excitation was direct observation of the morphologies of the trench structure. Figure 2c displays the time-resolved PEEM images recorded following excitation with p-polarized spatially separated femtosecond pulse pairs (The centers of the two spots are about 70 μm apart) with the various time delay (ΔT) between the pump and probe pulses (λ = 800 nm, the corresponding optical period is 2.67 fs). P-polarized pump pulse is used efficiently to excite the SPP and a clear static or dynamic interference pattern induced by the pump or a probe femtosecond laser and the laser-induced SPP could be observed. Thanks to the use of the few-cycle laser pulses (sub-10 fs) for the ITR-PEEM experiment, which has a short coherent superposition time of the probe laser pulse with the pump-excited SPP, we can intuitively capture a time-resolved image of the SPP field in the same probe region by adjusting the time delay (ΔT) between pump and probe pulses [24]. It should be mentioned that the zero delays are defined as the time in which SPP fields and probe pulses are overlapped in the edge of the probe region. According to the interference pattern shift in Fig. 2c, the group velocity of the SPP is calculated to be approximately 0.94c at a silver–vacuum interface for excitation with 800 nm light and this is consistent with the previous report [29].

Moreover, in the probe region, the interference patterns from a superposition of SPP and probe pulses belong to the three-photon photoemission (3PP) process (λ = 800 nm, 1.55 eV). Since the p-polarized probe pulse is incident at 65 degrees (as determined by the PEEM illumination configuration), 3PP signals are dominated by the transverse component of the interference field [30]. The electric field vectors diagram of the probe pulses and the SPP is given at the bottom in Fig. 2c. Figure 2d displays the PEEM image of the time delay ΔT = 13.33 fs using the excitation of p-polarized pulses and probed using s-polarized pulses. Similarly, the clear static interference pattern could be observed in the vicinity of the trench. Nevertheless, no dynamic interference pattern is observed in the probe region. This result is attributed to the fact that the electric fields of s-polarized light are orthogonal with the transverse and longitudinal components of the SPP, which causes no interference. The corresponding electric field vectors diagram of the probe pulses and the SPP is given at the bottom in Fig. 2d. The results show that this method of ITR-PEEM experiment of the incident light perpendicular to the long axis of the trench is not sensitive to the probing of the longitudinal component of SPP near-field.

Based on the above experimental results and the near-field superposition principle, the sensitivity of the s-polarized probe pulses to the longitudinal component can be improved by introducing a longitudinal component of SPP near-field on the x-axis component. Namely, the interference between s-polarized probe pulses and longitudinal component of SPP near-field can be realized by this method. Because the KL with KS satisfies Snell’s law of refraction [31, 32], this purpose can be realized in non-collinear mode. (The longitudinal component of the laser wave vector KL and the SPP k-vector are non-collinear).

Figure 3a shows the analytical calculation by the complex dielectric function of silver and FDTD simulation of the change of the propagation direction of SPP (α) with the angle (θ) of KL relative to the in-plane normal (the direction of the black dotted line) and the figures show that the calculation results (curve) are well fitted by the simulation (data points). This result implies that in non-collinear mode, the longitudinal component of SPP near-field will have a projection on the x-axis.

Fig. 3
figure 3

Analytical calculation by the complex dielectric function of silver and FDTD simulation of the change of the propagation direction of SPP (α) with the angle (θ) of KL between the in-plane normal (the black dotted line) (a) and they satisfy Snell’s law of refraction, as shown in the illustration. The near-field distribution of the longitudinal ((EX)2 and (EY)2) and transverse (EZ)2 components of the interference images of incident light and SPP field recorded following (b) p-polarized (c) and s-polarized irradiation and the direction of polarization are shown in the illustration. The white dotted frame indicates the position of the trench

Full size image

Furthermore, we used FDTD simulation to capture the distribution of the electric field near the surface of the silver film (x–o–y). The electric field monitor has been placed 5 nm above the silver film and it shows the near-field directly on the silver film. The near-field distribution of the longitudinal ((EX)2 and (EY)2) and transverse (EZ)2 components of the interference images formed between p-polarized (b) or s-polarized (c) irradiation (λ = 800 nm) with SPP field are shown in Fig. 3, respectively, and the direction of polarization is marked in the illustration. To map the longitudinal component of SPP, in the following experiments, the incident light is set along the long axis of the trench (θ = 90°). In Fig. 3b, the interference patterns were observed from the near-field of (EY)2 and (EZ)2 components, and the electric field intensity of the latter is much larger than that of the former. This result is consistent with the analytical calculation results based on the wave theory of electromagnetic field, i.e., the calculated ratio of the SPP electric field components (EST)2/(ESL)2 on the surface of the silver film is about 31(λ = 800 nm). In short, by comparison of the magnitudes of the longitudinal ((EX)2 and (EY)2) with transverse ((EZ)2) components, it is clear that the transverse component plays a primary role in the formation of the resultant interference pattern. Namely, despite the existence of the interference of the longitudinal components along the y-direction under p-polarization conditions, the contribution of the longitudinal components of SPP to the interference signals can be negligible. In such a case, the accurate capture of the transverse component information of the SPP near-field can be ensured. Moreover, no interference pattern in the electric field distribution of (EX)2 component can be seen in the left panel of Fig. 3b due to the absence of EX for p-polarized light. It needs to be mentioned that to better display the interference signal in Fig. 3, we have fine-tuned the value of the color bar and set different the starting values in the panels in Fig. 3.

Differently, in Fig. 3c, only the longitudinal component ((EX)2) has interference fringes, although the transverse ((EZ)2) component has a stronger electric field which is determined by the boundary conditions of the SPP [22]. This is caused by the interference pattern from the superposition of the x-component of SPP and the laser light under s-polarized irradiation (note different scales in Fig. 3b, c). The results demonstrate that the sensitivity of the s-polarized probe pulses to the longitudinal component can be improved in the non-collinear mode, i.e., method can be used to probe the longitudinal component of the SPP near-field. The above results suggest that spatiotemporal imaging of the transverse and longitudinal components of SPP near-field can be captured using the p-polarized and s-polarized probe pulses, respectively, with the non-collinear trench excitation mode in the ITR-PEEM experiment. It should be mentioned that the laser pulses of 80 fs are used in this simulation to ensure that the angular dispersion in the injected wave vector is minimized [33]. The simulation using 80 fs pulse is suitable for explaining our experimental results.

To experimentally implement the above scheme, the schematic illustration of the ITR-PEEM experiment of the incident light along the long axis of the trench (θ = 90°) be designed, as shown in Fig. 4a. A trench of size 20 × 1μm2 was used in the following experiment to avoid the influence of ring-like interference pattern 11 excited by the end of the trench on the ITR-PEEM experiment.

Fig. 4
figure 4

Schematic illustration of ITR-PEEM experiment of the incident light along the long axis of the trench (θ = 90°) (a) and PEEM images of the spatially separated ITR-PEEM experiment of the incident light along the long axis of the trench (θ = 90°) at 20 × 1μm2 trench structure. The pump pulse with a polarization angle of 45° off the sample surface as indicated in the insert (c) is used efficiently to excite the SPP, and the p- or s-polarized probe pulse interrogates the SPP at a remote point. PEEM image obtained with a pump of 45 degree-polarized laser pulses and probe of p-polarized pulses (b, c, d) and s-polarized pulses (e, f, g) under different delays, respectively. The dynamic interference pattern is derived from the superposition of the laser and SPP electric field in the probe region. The vectors diagram of the laser field and the SPP field is given at the bottom in (d, g). A white dashed rectangle is used to mark the location of the trench in (b), and yellow dashed ovals represent the approximate positions of the pump and probe beams. Yellow arrows indicate the direction of propagation of SPP. Zero delays are defined as the delays in which SPP of the pump launching and probe pulses of s-polarized are overlapped in the edge of the probe region, as shown in Fig. 3e

Full size image

Figure 4b, c, d shows the time-resolved PEEM images recorded following excitation with spatially separated femtosecond pulse pairs at 20 × 1μm2 trench structure. The pump pulse with a polarization angle of 45° off the sample surface as indicated in the insert (c) is used efficiently to excite the SPP [33], and the p-polarized probe pulse (the power is about 60 mW) interrogates the SPP at a remote point. The results display that the SPP dynamically propagates with the pump-probe pulse time delay. From vectors diagrams of the laser field (p-polarized) and the SPP field that are given at the bottom in figure (d), it is noted that there could be photoelectron contributing from the interference between the component ELY of the p-polarized probe light and the projection of ESL along the y-axis. One might question the feasibility of independently probing the transverse component of SPP in the current scheme. However, as discussed in Fig. 3, the photoelectron contribution from the longitudinal component of SPP can be negligible when one considers the large difference in field intensity between the transverse and longitudinal components of SPP. Due to the dynamic interference patterns in the probe region in (b, c, d) are dominated by the transverse field component of SPP and the probe pulses, thus reflecting the spatiotemporal information of the transverse component of the field of SPP. This demonstrates the spatiotemporal imaging of the transverse component of SPP near-field can be implemented using the ITR-PEEM experiment of the incident light along the long axis of the trench under the p-polarized probe pulses. It is worth mentioning that the zero delays are defined as the delays in which SPP of the pump launching and probe pulses of s-polarized are overlapped in the edge of the probe region, as shown in Fig. 4e, the interference fringes only appear near the SPP envelope peak ensure a more preciously zero delay when the s-polarized pulse is employed, and this will be discussed in detail below.

Figure 4e, f, g displays the time-resolved PEEM image of the same delay time but is recorded with an s-polarized probe pulse. To capture a clear interference pattern, the maximum power (~ 160mW) that the laser can provide is used to the s-polarized probe pulses. The s-polarized probe laser field exists only in the plane of the sample, unlike Fig. 2a, the observable interference patterns in the probe region of Fig. 4e, f, g results from the superposition of the longitudinal component of SPP near-field and the laser light. The vectors diagram of the laser field and the SPP field is given at the bottom in Fig. 4d, g. This result confirms that the spatiotemporal information of the longitudinal component of SPP near-field can be captured using the ITR-PEEM experiment in the non-collinear trench excitation mode under the s-polarized probe pulses. To sum up, the spatiotemporal imaging of the respective components of SPP near-fields in nano-femtosecond scale can be simultaneously realized by the ITR-PEEM experiment in the non-collinear trench excitation. It should be mentioned that the power of the pump pulse is the same as the s- and p-polarization probe in the experiment. To observe clear interference patterns, the contrast of the images was adjusted for the case of the s-polarization probe.

Furthermore, to compare the photoelectron (PE) intensity of the two interference signals, the PE intensities of the interference patterns in the white box in Fig. 4c, f are obtained by subtracting the PE signal background extracted from the laser-irradiated region. It is found that YT/YL is about 7.9. Here, YT and YL, respectively, represent PE intensities of interference patterns from SPP and incident laser electric field transverse and longitudinal components. The result shows that the PE intensity induced by the interference of SPP and incident light longitudinal field (f) is far lower than that induced by the interference of the transverse field (c).

We attribute the above results into three aspects. (i) The difference between s- and p-polarized light reflection coefficients of silver films. We considered the absorption and reflection coefficients of the silver film. Based on Fresnel’s law, for a light of 800 nm wavelength and oblique incidence 65°, the reflectance of p-polarized light and s-polarized light are 0.942 and 0.988. Namely, in the experiment, the effective power of p-polarized probe pulse (the power is about 80mW) and s-polarized probe pulse (the power is about 160mW) are 3.48mW and 1.92mW, respectively, and this is a significant difference when considering that the 3-photon photoemission process is involved. (ii) The interference between the s-polarized probe pulse and part of the SPP longitudinal components. Specifically, in figure (f), the angle between SPP propagation direction and x-axis is 62.7 degrees, as shown in Fig. 3a, and only the projection of the longitudinal component of SPP (EX = 0.46 ESL) onto the x-axis satisfy the conditions for interference with the s-polarized probe light vector. (iii) According to the simulation results and analytical calculations in Fig. 3b, c, the transverse component of the SPP electric field on the surface of the silver film is about 5.5 times that of the longitudinal component. Therefore, we speculate that the above three factors lead to the PE intensity induced by the interference of SPP and incident light longitudinal field (f) is far lower than that induced by the interference of the transverse field (c). As a result, the number of interference fringes when s-polarized probe pulses are used (figure e–g) is much less than that when p-polarized probe pulses are used (figure b–d). Correspondingly the interference fringes only appear near the SPP envelope peak, as marked by dash line rectangles, in Fig. 4e–g in the probe region for the s-polarization probe mode, and almost all the information of the entire envelope of the SPP is revealed for p-polarized probe mode (Fig. 4b–d). The above discussions explain the interference pattern that located in the lower right half of the ellipse in figure (b) can still be observed.

Based on the above discussion, we evaluated the ratio between transverse and longitudinal electric field components. First, we measured the contrast of the interference patterns from Fig. 4c, f, respectively:

$$\frac{{Y_{PMAX} }}{{Y_{PMIN} }} = 3.46$$
(1)
$$\frac{{Y_{SMAX} }}{{Y_{SMIN} }} = 1.42$$
(2)

Here, YP and YS represent the electronic yield extracted from Fig. 3c, f, respectively. Taking into account the three-photon process of photoelectron emission, the contrast of the interference fields of the probe area from Fig. 4c, f can be expressed as,

$$\left( {\frac{{\sin 65^\circ E_{P} + E_{ST} }}{{\sin 65^\circ E_{P} - E_{ST} }}} \right)^{2} = 1.51$$
(3)
$$\left( {\frac{{E_{S} + \cos 62.7^\circ E_{SL} }}{{E_{S} - \cos 62.7^\circ E_{SL} }}} \right)^{2} = 1.126$$
(4)

And,

$$\left( {\frac{{E_{P} }}{{E_{S} }}} \right)^{2} = 1.82$$
(5)

Here EP and ES are the effective electric field strength of the p and s polarization probe light on the surface of the silver film, respectively. Then, the ratio between transverse and longitudinal electric field components is,

$$\frac{{E_{ST} }}{{E_{SL} }} = 2.09$$
(6)

It is found that the above result is lower than the analytical calculation result (EST/ESL = 5.56). We attribute the above difference to the fact that when the probe pulse is s-polarization, the presence of the transverse electric field component of SPP will form a stronger background signal in the PEEM image, which reduces the contrast of the interference pattern [22], as shown in the figure (e–g) as shown. As a result, the ratio between transverse and longitudinal electric field components is smaller than the analytical calculation.

Compared to our previous work (reference 24), several advances have been made. The use of the few-cycle laser pulses (sub-10 fs) greatly reduces the coherent superposition time of the probe laser pulse with the pump-excited SPP. The dynamic propagation of the femtosecond SPP pulses in the same probe region is captured by adjusting the time delay between pump and probe pulses rather than simultaneously changing the position of the probe light and the delay time of the pump-probe pulse in each capture in the case of 130 fs laser pulse, which hinders the capture of the spatiotemporal information of the SPP near-field in nano-femto scale. Moreover, the contributions of the transverse and longitudinal components of SPP near-fields to interference patterns were systematically analyzed by the calculation using the complex dielectric function of silver and FDTD simulation under the incident light along the long axis of the trench. The above results confirm the feasibility of the spatiotemporal imaging scheme of the longitudinal and transverse components of the SPP near-field in this work. Thanks to the few-cycle laser pulses, we can realize independent spatiotemporal imaging of the transverse and longitudinal components of SPP near-fields on a nano-femtosecond scale. Finally, the ratio between transverse and longitudinal electric field components is measured and analyzed for the first time using ITR-PEEM.

4 Conclusions

In summary, we performed the spatiotemporal imaging of the few-cycle SPP in nano-femtosecond scale using two different excitation schemes that include the incident light along and vertical to the long axis of the trench by oblique incidence ITR-PEEM. The intuitively dynamic propagation of the transverse and longitudinal components evolution of the identical SPP field in the same probe region was captured using few-cycle (sub-10 fs) pulses, via utilization of p-polarized and s-polarized probe pulses, respectively, under the ITR-PEEM experimental scheme of the incident light along the long axis of the trench. The feasibility of the spatiotemporal imaging scheme of the longitudinal and transverse components of the SPP near-field is conformed using the complex dielectric function of silver and FDTD simulation. Meanwhile, it is found that the photoemission yields induced by the longitudinal field are far lower than those induced by the transverse field, and the interference patterns induced by s-polarized probe pulses and detectable longitudinal field of SPP is concentrated near the peak of the SPP envelope. The factors that induce weak fringe intensity in the longitudinal probe scheme are semi-quantitatively analyzed and, the ratio between transverse and longitudinal electric field components is measured for the first time using ITR-PEEM. In short, this work provides a basis for further application of the few-cycle SPP and it is the premise of mapping the 3D SPP space–time field using the pump-probe scheme.