Interface Charge Characteristics in Polymer Dielectric Contacts: Analysis of Acoustic Approach and Probe Microscopy

Interfaces are essential components in polymer contact systems, which widely exist in electronic devices and power equipment. Interface charge originating from the mismatch of the electronic structure in interfaces is one of the key issues to modify the device performance due to its multifunctional migration and accumulation behaviors. Hence, the detection and analysis of the interface charge characteristics are of great importance to deeply understand the polymer contact system in various devices. This paper presents an overview of recent research progress in the interface charge properties at dielectric interfaces. Based on the theoretical analysis of the MWS polarization and electronic localized states, two typical approaches of discussing the interface charges from micrometer to millimeter are mainly studied. Acoustic method is prevalent in detecting the space charge in various dielectrics. However, owing to its limited resolution (several µm), it is difficult to clarify the charge distributions at the interface with a micro‐scale. Probe microscopy presents a promising technique due to its flexible surface potential detection at a submicron scale. The challenges and prospects of acoustic and probe microscopy methods are discussed in this paper. The advanced techniques of interface charges can promote the development of new energy electronic devices.

Interfaces are essential components in polymer contact systems, which widely exist in electronic devices and power equipment. Interface charge originating from the mismatch of the electronic structure in interfaces is one of the key issues to modify the device performance due to its multifunctional migration and accumulation behaviors. Hence, the detection and analysis of the interface charge characteristics are of great importance to deeply understand the polymer contact system in various devices. This paper presents an overview of recent research progress in the interface charge properties at dielectric interfaces. Based on the theoretical analysis of the MWS polarization and electronic localized states, two typical approaches of discussing the interface charges from micrometer to millimeter are mainly studied. Acoustic method is prevalent in detecting the space charge in various dielectrics. However, owing to its limited resolution (several µm), it is difficult to clarify the charge distributions at the interface with a micro-scale. Probe microscopy presents a promising technique due to its flexible surface potential detection at a submicron scale. The challenges and prospects of acoustic and probe microscopy methods are discussed in this paper. The advanced techniques of interface charges can promote the development of new energy electronic devices. and moisture, [68] can affect the interface charges at polymer/ polymer dielectric contacts. Furthermore, changing the electrode material and surface conditions can be useful in detecting interface charges at various electrode/dielectric structures. [69][70][71][72] However, acoustic methods have limitations for measuring interface charges at a submicron scale, such as less than 1 µm, due to the low resolution of the sensor. [56,73,74] In recent years, the probe microscopy method has enabled atomic-level resolution imaging, making it possible to study charge behaviors at the submicron scale. [75][76][77][78][79] Kelvin probe force microscopy (KPFM) is a technique used to measure the potential difference between a sample and the tip. [79,80] KPFM measures the work function between the tip and the surface as a function of potential difference. [81,82] This technique is beneficial for studying electronic properties near the surface in nanotechnology and semiconductor fields. [83] In addition, the probe microscopy technique is effective in detecting interface charge on a microscopic scale. [84][85][86][87] KPFM with a direct current offset voltage can be used to study the interface charge properties between the nanofiller and polymer matrix in nanocomposites, contributing to understanding the effect of the interface (≈10 nm) on the performance of nanocomposites. Unfortunately, a correlation between surface potential and interface charges at the submicron scale is still lacking.
Literature results have recently summarized the behaviors of polymer/polymer interfaces, [16,32] electrode/polymer interfaces, [21,88,89] and filler/matrix interfaces. [3,90] However, there is still a lack of synthetic review on the interface charge characteristics. This paper focuses on the recent progress of interface charges in polymer dielectric contacts, with a particular emphasis on the analysis of acoustic approach and probe microscopy methods. First, we analyze the classical MWS polarization theory and localized states theory, as well as the classification of interfaces. Second, we introduce the characterization techniques of interface charge by acoustic and probe microscopy methods, followed by the interface charge properties of various interfaces. Finally, we briefly summarize the recent works and propose future opportunities and challenges for interface charges in polymer dielectric systems.

Origin of Interface Charges
Interfaces with complicated properties, originating from material structure, functions, and applied conditions, dominate the electrical performance of the dielectric system in comparison to the bulk material. [3,16,37,91] At the macroscopic level, the interface should be a dimensionless domain that is different from the attached materials, exhibiting a significant change in parameters across the interface. [90] As a result, interface charges remain due to the change in material structure. [16] These interface charges are a type of static charge that can be regarded as a uniform accumulation, [54] which can affect the electric field, polarization, and charge conduction in multilayered structures. [54,92,93] At the microscopic level, the interface is a transition region ranging from a few nanometers to micrometers. [94][95][96] Therefore, it is challenging to use a uniform parameter to study interface behaviors. The interface can be a domain that depends on the attached material structure and www.advmatinterfaces.de defects. The discontinuity of the material structure can create a thin region where physical and chemical parameters exhibit transition distributions. The interface charge is affected by the electronic states and impurity energy levels. [97][98][99] This section describes the types of interfaces and the definition of interface charge, followed by the principles of MWS polarization and localized states.

Interface Patterns and Related Charges
An interface is defined as the region where two materials connect and interact with each other. [91,100] This interaction is largely dependent on the attached two-material phase, [90] and the various patterns of the interface are determined by the properties of the materials involved. Interfaces are usually formed under the influence of applied stresses, such as mechanical, electric, and thermal forces. Table 1 summaries an overview of the typical interface patterns observed in solid polymer systems and the associated interface charges. The three types of interfaces studied include 1) polymer/polymer interface, for instance, low-density polyethylene (LDPE)/ EPDM, and XLPE/XLPE double-layered dielectric systems; 2) electrode/polymer interface, such as Al, Au, and semiconductive (SC) electrodes/XLPE; and 3) polymer/filler interfaces, such as epoxy resin (EP)/graphene oxide (GO), LDPE/ SiO 2 , XLPE/boron nitride nanosheets (BNNS), EP/SiC and XLPE/Al 2 SiO 5 composites. Under alternating current (AC) stress, the electric field distribution in the insulating material depends on its dielectric permittivity, [101] while under direct current (DC) electric field, it is determined by the material conductivity. [102] In polymer contact systems, interface charges can be attributed to interfacial polarization, charge injection, and trapping/de-trapping. [54,[103][104][105] Hence, interface charges are commonly observed in multi-layered materials, as shown in Table 1.
In recent years, researchers have focused on establishing a correlation between interfacial regions and material performance by understanding the accumulation of interface charges. [3][4][5]90] Figure 2 illustrates the interface charges formed by electrode injection, interfacial polarization, and localized state trapping in the three types of interface. [112][113][114] The accumulated interface charges affect various material properties such as electric field distribution, [16] energy level alignment, [98,115,116] carrier migration, [117] thermal conductivity, [118] and dielectric loss. [119,120] In an electrode/dielectric/electrode structure, the current continuity principle suggests that when the net current flux near the electrode is higher than the bulk current flux, the charges can accumulate at the electrode/polymer interface. Consequently, the homocharge accumulation leads to a decrease in the electric field near the electrode interface. [32] The net current flux near the electrode reduces to keep the equilibrium with the bulk current flux. [32,67]

Effect of Interfacial Polarization
Interfacial polarization is produced by the separation of moving positive and negative charges under an applied field, which forms positive and negative space charges at the interfaces between different dielectrics. [121] Figure 3a depicts the basic assumptions of the MWS interfacial polarization model. The voltage on the double-layered dielectric is distributed according to the capacitance (Figure 3b). The mismatch of conductivity and dielectric constant results in the build-up of interface charge (Figure 3c). Using the dielectric constants (ε 1 and ε 2 ), conductivity (κ 1 and κ 2 ), and thickness (d 1 and d 2 ) of each layer, the voltage (V 1 (t) and V 2 (t)), relaxation time τ, and the interface charge density σ can be calculated by [16,41,122,123]  EP/SiC Electrode injection/ Interfacial polarization Interface enhances thermal conductivity [110] XLPE/Al 2 SiO 5 Electrode injection/ Interfacial polarization Interface charge causes a peak in dielectric loss [111] Adv. Mater. Interfaces 2023, 10, 2300087 The MWS polarization model is useful for studying space charge behavior at the interface in heterogeneous materials. [16,40,41,124] The advantage of this model is to use macroscopic parameters (such as the dielectric constant and conductivity of the material) to discuss the interface charge, without using assumed values of trap depth and activation energy related to carrier injection and transport processes. [32,41,125,126] However, it should be noted that the MWS polarization model suggests that there should be no charge at the interface between the same materials, which contradicts experimental results. [16,32] It is commonly believed that the charge injection from the electrode, ion dissociation, and impurities in the material affect the accumulation of interface charges. [49] As a result, the measured results of interface charges may deviate from the MWS theoretical prediction.

Effect of Localized States at Interface
The periodic potential field of a solid is abruptly interrupted and distorted at the interface, [127] requiring the atomic arrangement to be adjusted to maintain system stability. As a result, additional electronic states, such as localized states could be formed. [128] The localized states increase the trap density and energy level at the dielectric interface. [65] It depends on the electronic structure and interfacial morphology of the material. As a kind of localized state, charge traps are particularly important in either enhancing or restricting the charge transport of dielectric materials, [129] At a mesoscopic scale, the electronic structure is complex and closely related to the dielectric materials. [127,130,131] The electronic structure at the interface includes the energy level arrangement at a thin scale and the band bending in the thicker region. [132] The former contributes to carrier injection (e.g., electroluminescent devices or spectral sensitization) and the latter causes carrier separation (e.g., solar cells). [98] Interface charge accumulation occurs due to the complex localized states and the introduced potential barrier at the interface. [16] The study of electronic structure is significant in understanding space charge behaviors at the interface. Figure 4a illustrates the XLPE/EPDM double-layered dielectric structure and the electron trap distributions within the dielectrics. [54] The horizontal axis of the electron trap  www.advmatinterfaces.de distribution diagram represents trap energy, while the vertical axis represents trap density. The electron trap distribution diagram reveals that XLPE mainly contains shallow traps, while EPDM contains shallow and deep traps. Under the application of a positive voltage, the deep trap present in EPDM immediately captures the electrons migrating from the XLPE side to the interface, causing negative interface charge accumulation ( Figure 4b). Interface charge trapping plays an important role in charge dynamics, which can be used to analyze the polarity of interface charge and explain the bipolar charge behavior at the interface. The interface structure formed by the POSS and polypropylene (PP) and the corresponding localized states of the interface region is shown in Figure 4c. [133] The contact between the filler and the matrix introduces trap energy levels at the interface. Figure 4d shows the charge transport process related to traps. The green lines represent the trap sites, and the red lines represent the charge transport, trapping, and de-trapping process. Charges injected from the electrode can be trapped by the traps near the electrode/dielectric interface, forming the accumulation of homocharges. [134,135] Moreover, charges need to overcome a potential barrier after being trapped. The carriers trapped by the deeper traps introduced by the filler/matrix interface are difficult to de-trap, thus inhibiting the injection and migration of charge. [133,[136][137][138]

Experimental Techniques of Interface Charges
Over the past decades, several methods for measuring interface charges in polymer dielectric systems have been proposed and studied, including optical, thermal, acoustic, and probe microscopy techniques, as shown in Table 2. Optical methods, such as the Pockels and the Kerr effect, are suitable for testing charges in thin solid samples and liquids, respectively. [11,139] Thermal methods, comprising the thermal pulse method (TPM), thermal step method (TSM), and laser intensity modulation method (LIMM), are also used for measuring charges. [56,140] Acoustic methods, such as the PEA and PWP methods, have been used extensively. [56,141] In particular, the PEA method is widely used in space charge measurements in various polymer dielectric systems. [56,59,142] However, PEA has a limitation of low space resolution in thin films due to the response of piezoelectric sensors and pressure waves. [57,77] Additionally, KPFM and electrostatic force microscopy (EFM) based on flexible scanning methods are promising technologies for measuring nanoscale surface and interface charges. [75,85,86,94] This section mainly discusses the macroscopic technique of the acoustic method (PEA) and the microscopic technique of the probe microscopy method (KPFM).  [54] Copyright 2019, IEEE. (c,d) Reproduced with permission. [133] Copyright 2019, Elsevier. www.advmatinterfaces.de

Acoustic Method
This section focuses on the widely used PEA method for the measurement of interface charge. The basic principle of space charge measurement is that when a pulsed voltage with superimposed DC stress is applied to a laminated sample, the applied pulsed electrical stress leads to mechanical vibrations of the space charge in the sample and changes in the Coulomb force distribution, which result in the generation of pressure waves. [56,57,60] The pressure wave propagates toward the lower electrode and is converted to a voltage signal by a piezoelectric transducer. The output signal is then amplified and observed using an oscilloscope. The principle of the PEA method for interface charge measurement is illustrated in Figure 5a.
As shown in Figure 5b, the voltage signal output by the piezo-electric sensor contains both bulk and interface charge information, but the real output signal deviates from the ideal output signal. In order to obtain accurate interface charge characteristics, researchers have made many improvements in signal processing technology. For example, the influence of temperature on acoustic velocity, attenuation coefficient, and dispersion coefficient is considered. [152][153][154] By obtaining the reference signal at a certain temperature, the relationship between the attenuation coefficient and temperature can be established. [153] The reference signals at different temperatures can then be reconstructed to recover the space charge signals at high temperatures. Moreover, it is necessary to improve the transfer function to suppress the waveform distortion under a temperature gradient. [63] The sound velocity gradient, acoustic attenuation, and dispersion caused by a temperature gradient have been studied to improve the voltage-charge calibration coefficient, attenuation, and dispersion coefficients in the double-layered dielectrics. To calibrate the attenuation and dispersion of the sound wave in the coaxial cable, the system transfer function and the acoustic wave attenuation function have been analyzed, which is useful in recovering the space charge waveform of the coaxial structure sample. [152] Figure 5c shows the acquisition method of time and spatial resolution of the PEA method. Time resolution is related to the half-width of the output signal, whereas spatial resolution depends on the half-width and speed of sound in the dielectric. [155] Researchers have made several improvements to improve the resolution of the PEA device and satisfy different testing conditions. Due to the limited frequency bandwidth of the lead zirconate titanate (PZT), high-frequency components of the pulse acoustic signal are lost when it passes through the PZT, resulting in a series of oscillations in the output signal. [60,156] Therefore, it is necessary to avoid the loss of highfrequency components by using a deconvolution procedure to achieve a resolution of nearly 200 µm. [155] Later, polyvinylidene fluoride (PVDF) piezoelectric films were used instead of PZT to solve the loss of high-frequency components, which improved the resolution of PEA method to 60 µm. [60] Compared with the traditional measurement of space charge under DC voltage, the measurement of space charge under AC voltage requires a pulse power supply with a narrower pulse width and higher frequency, and a thinner piezo-electric sensor film. With an improved pulse source having an amplitude of 700 V, a full width of 4.6 ns, and a maximum repetition rate of 10 kHz, multiple space charge distribution measurements can be completed in a single AC voltage cycle. [157] Moreover, precise phase  [60,143] Fast test speed/ High noise immunity/ Simplicity of the measurement cell/ Non-destructive [16] Low resolution [31,77] / Acoustic signal needs calibration [63,69] PWP Fast test speed/ High signal-to-noise ratio/ Non-destructive [144] Low resolution/ Pressure wave signal needs calibration [143] Optical Pockels effect Calculation of electric field using the change of optical refractive index to deduce charge signal [11,139] High spatial resolution/ Suitable for solid dielectric films [141,145] Destructive to the sample [139] Kerr effect High spatial resolution/ Suitable for liquids [146] Thermal TPM Thermal diffusion disturbs charge and induces electrical signal [56,140] Non-destructive/ Suitable for thin samples [147] Low resolution/ Sensitivity and resolution are strongly related to spatial position TSM Non-destructive/ Suitable for thick samples [148] LIMM Three-dimensional electric field distribution can be provided Probe microscopy KPFM By compensating or minimizing the electrostatic force between the probe and the sample [84] Non-destructive/ Nanoscale resolution/ Direct measurement of surface potential changes caused by charges [149] Probe needs calibration/ Affected by test environment and sample surface condition [150] EFM By measuring the electrostatic force between the tip and the sample [95] Nanoscale resolution/ Ultra-high precision [151] www.advmatinterfaces.de control can be achieved by combining the multi-channel function generator.

Probe Microscopy Method
The acoustic method is useful to characterize the interface charges in the multi-layered contact structures in power equipment, [7,43,49,158] especially high-voltage power devices. In this case, the system has obvious interfaces, and it is reasonable to ignore the interface region (without observed thickness). Hence, it is simple to use the acoustic method to calculate interface charges with a relatively high resolution (≥1 µm). [56,60,159] However, in the field of microelectronics and nanocomposites, interfaces are not identifiable due to their small scale (nanometer). For example, the morphology, structure, composition, and molecular bonding in the interfacial region between the filler and polymer matrix are complex, making it challenging to understand the influence of interfacial properties on composites. [3,75] The traditional acoustic method is not suitable for testing space charges at the interface region due to limited resolution. [75][76][77]85,86,94] Acoustic method calculates the charge density with a depth resolution of up to about 1 µm and typically has no lateral resolution. [160] The depth information provided by the acoustic method is not sufficient for accurately studying interfacial phenomena in thin dielectric films (less than 1 µm thick). [160] Therefore, advanced microtechniques for interface charge evaluation should be used. Recently, probe microscopy methods, especially KPFM, has provided a new direction for studying interfacial phenomena with a microscale. [87,[136][137][138] Based on atomic force microscopy (AFM) technology, KPFM quantifies the contact potential difference (CPD) between the probe and the sample surface by detecting the electrostatic force of the capacitance. [75,94,95,[161][162][163] The first report on KPFM can be traced back to 1898. [164] KPFM can be divided into closedloop mode and open-loop mode according to the difference in surface potential difference acquisition methods. The former eliminates the electric field between the probe and the sample by compensating the voltage, [165] while the latter uses the vibration state parameters (amplitude or frequency) of the probe. [80] However, the possible oscillation of the DC compensation feedback loop in the closed-loop mode affects the stability of the measurement system. Therefore, the open-loop mode is more widely used. Additionally, based on the modulation mode, KPFM can be classified into amplitude modulation (AM) mode and frequency modulation (FM) mode. [166] AM can be used to quickly inspect large surfaces, and the required AC drive voltage is generally low. While AM mode is stable and easy to realize, the geometry of the probe tip and cantilever leads to high stray capacitance, which limits the resolution. [167] In contrast, FM mode is sensitive to electrostatic force gradient, making it more suitable for surface potential resolution, but it is easily affected by the surface roughness of the sample. Figure 6 illustrates the KPFM test configuration and electrostatic interaction between the probe tip and sample surface (the potential difference between the tip and sample surface is compensated by U DC ). First, KPFM performs tap mode imaging to obtain the sample surface morphology. Second, topographic information is used to retrace the baseline, and the probe is placed at a given lift height of z above the sample surface. [168] At the same time, an external voltage is applied to the tip to excite the electrostatic interaction, and the potential difference between the tip and the sample can be expressed as [94] sin ts dc CPD a c ω ( ) where V dc and V ac sin(ωt) are the DC and AC voltages applied to the tip, and V CPD represents the surface potential difference between the tip and the sample. Ignoring the surface electrostatic charge, the electrostatic force on the tip can be expressed as Figure 6. Schematic diagram of KPFM test configuration and electrostatic interaction between the probe tip and sample surface (compensation of surface potential difference V CPD between probe and sample by DC voltage U dc ). Reproduced under terms of the CC-BY license. [169] Copyright 2010, The Authors, published by IOP Publishing Ltd. www.advmatinterfaces.de where z is the lift height of the tip, and C is the capacitance between the probe/cantilever and the sample. Then, the surface potential difference V CPD can be calculated as where A ω and A 2ω are the vibration amplitude of the probe at the fundamental and second harmonic frequencies of V ac , respectively. Then, the surface potential extracted from V CPD can be expressed as where φ tip and φ sample are the surface potential of the tip and the sample, respectively. Then the electric field E(x) and the charge distribution ρ(x) can be calculated as [85,86] The resolution of KPFM surface potential measurements can be influenced by several factors, including probe geometry, material, parasitic effects (e.g., capacitive coupling), sample surface conditions, and test environment. [150] To achieve accurate surface potential measurements, researchers have implemented various improvements: 1) Calibration methods can be used to determine precise tip work functions. For instance, multiple linear fitting of the contact potential difference between the sample and the tip can reduce the uncertainty of the tip work function. [170] 2) Ambient humidity and the sample surface state (e.g., surface reaction/oxidation or other pollutants) can diminish measurement accuracy, necessitating the establishment of stable and artifact-free imaging conditions. For example, AM mode can increase the distance dependence of modulated electrostatic force, improve the sensitivity of short-range force, increase the contribution of the needle tip in surface potential measurement through the heterodyne method (frequency conversion technology), and reduce the surface potential measurement crosstalk caused by surface topography feedback. [167] 3) Optimize KPFM scanning speed. Reconstruction based on real-time force can be achieved by deconvoluting the cantilever transfer function, significantly improving the scanning speed of conventional KPFM (<20 µs). [171] In addition, test conditions can affect the conversion of surface potential into charge density. [172,173] In the typical dual-channel KPFM measurement process, tap mode topography scanning can trigger contact electrification between the probe and the sample, causing significant errors in surface potential measurement. [172] Reasonable decoupling of contact electrification and surface potential measurement can be achieved by decreasing the free amplitude, increasing the set-point amplitude, and using a probe with a lower spring constant.

Interface Charge Characteristics
This section presents various experimental results of interface charges analyzed by acoustic and probe microscopy methods in different dielectric structures. The former mainly focuses on the PEA technique, which has been widely used in the power cable, [174] transformer, [25,26,56,175] and gas-insulated switchgear insulating materials and structures. [176,177] To guide engineering applications, the interface charge characteristics that depend on external factors, such as temperature, electric field, moisture, and electrode condition, are described and discussed. The latter employs advanced microscopy of scanning techniques to obtain detailed information on material morphology and structures. In particular, the surface and interface charges by KPFM technique can present microcosmic parameters of interface phase (region), such as surface potential, charging, interface thickness, work function, injection barrier et al. [80,[178][179][180] It can directly measure the potential signal in the interfacial region and then analyze the charge injection and transport properties.

Effect of Temperature
Temperature plays a crucial role in space charge behaviors at interfaces in multi-layered materials. High-voltage power equipment, such as oil-immersed transformers, may reach high temperatures under operating conditions. [19,25,181] Additionally, the insulating film in superconducting cables, such as polypropylene lamination paper, polyimide, and kraft paper, must operate at low temperatures close to −196 °C. [182] Temperature can affect charge injection, transport, conduction, recombination, polarization, and trapping/de-trapping processes. [183,184] Therefore, it is necessary to measure and discuss the effect of temperature on the interface charge.
According to the MWS polarization model, the interface charge is affected by the conductivity and dielectric constant of the dielectric. [16,32,41] Considering that the conductivities of XLPE and silicone rubber (SIR) increase with temperature (Figure 7a), [7] there is a certain temperature at which the proportion coefficient K becomes greater than 1, resulting in a positive polarity of the interface charge (Figure 7b). The experimental results shown in Figure 7c conform to the MWS polarization model. With the increase in temperature, the interface charge density first decreases and then increases, demonstrating the temperature dependence of the interface charge polarity. Furthermore, enhanced charge injection at high temperatures also affects the polarity of the interface charge. [185] Figure 7d illustrates the charge dynamics in the XLPE/rubber double-layered dielectric at temperatures of 20, 40, and 70 °C. [123] At low electric fields (<10 kV mm −1 ), positive charges accumulate in the interface region (dotted line region in Figure 7d) at 20 and 40 °C. However, when the temperature increases to 70 °C, the polarity of the interface charge changes from positive to negative. According to the Poole-Frenkel effect, the high temperature reduces the potential barrier at the electrode/dielectric interface, enhancing charge injection through thermal excitation. [121] The interface charge build-up is attributed to electrons on the www.advmatinterfaces.de XLPE side that easily migrate and are captured near the interface by deep traps in the rubber. Notably, charge injection at a high field is dominant, [186] which may be responsible for no change in the polarity of the interface charge with the increase of temperature. A large number of electrons injected at the XLPE side rapidly neutralize the positive polarized charge, and the amount of negative charge accumulated at the interface increases with the increase of temperature.
Usually, the temperature in dielectric structures is not uniform, resulting in a distributed temperature gradient that affects the charge and electric field distribution. [63,152,187,188] Hence, it is necessary to study the influence of temperature gradient on interface charges. In the case of the oil-pressboard double-layered dielectric, negative charges moved in the oil layer and accumulated at the interface between the oil and pressboard under isothermal conditions. [181] This indicates that the conductivity of oil is higher than that of pressboard paper at 20 °C. The polarity of the interface charge was consistent with that of the oil-side electrode. At a temperature gradient of 40 °C (20 °C for the oil and 60 °C for the pressboard), significant positive charges are injected from the high-temperature anode, and bipolar charges appear near the interface. In the presence of a temperature gradient, the higher temperature at the anode reduces the charge injection threshold field and increases the charge carrier mobility. In addition, the temperature gradient causes the nonuniform distribution of conductivity, [126] and the conductivity of the dielectric at the high-temperature side may increase faster with temperature. Injected charge carriers migrate toward the interface and recombine with the negative charges, resulting in a decrease in negative interface charges. Some extra positive charges pass through the interface barrier and form a positive charge accumulation at the oil side near the interface.

Effect of Electric Field
Apart from temperature, the electric field also affects charge injection, conduction, recombination, polarization, and trapping/de-trapping processes. [66,187,189,190] Figure 8 illustrates the space charge characteristics and electric field distribution of XLPE/EPDM double-layered structure (Figure 8b) under different DC voltages. [54] Notably, Figure 8a shows the time dependence of the interface charge polarity under an electric field of −20 kV mm −1 . Initially, a small quantity of positive charges appears at the interface, which then turns into negative charges. The polarity of the interface charge is affected by the interface polarization effect, wherein low dielectric (a-c) Reproduced with permission. [7] Copyright 2019, Springer Nature. (d) Reproduced with permission. [123] Copyright 2015, IEEE.

www.advmatinterfaces.de
constant and high conductivity contribute to the charge polarity. The appearance of a small number of positive charges may be attributed to the higher conductivity of EPDM than XLPE at a low electric field, and the polarity of interfacial polarized charges is temporarily dominated by the EPDM side. As the electric field increases, more positive charges are trapped by the deep traps on the EPDM side. Figure 8c demonstrates that the presence of interface charges causes a severe distortion of the electric field in the dielectric. For example, at an electric field of −40 kV mm −1 , a highly distorted electric field of up to 59 kV mm −1 appears in the EPDM layer.
Polarity reversal voltage is a phenomenon in the line-commutated converter-HVDC transmission system. [191] The voltage polarity can change from one polarity to another in a very short time. This sharp change of voltage will significantly enhance space charge accumulation at the interface and lead to electric field distortion and insulation aging. [41,192] Figure 9a illustrates the interface charge behaviors in an EPDM/LDPE double-layered dielectric before and after voltage polarity reversal. [41] The polarity of the interface charge is consistent with the polarity of the LDPE electrode side after a polarity reversal. However, during voltage polarity reversal, bipolar charge accumulation and dissipation occur at the interface. Notably, a positive charge packet is formed, moving toward the negative charges at the interface. When the interface positive charges transfer to negative charges during voltage polarity reversal, the positive charge packet moves quickly toward the cathode. Consequently, as demonstrated in Figure 9b, the complex interface charges lead to severe electric field distortion in the dielectric, and the maximum electric field inside the EPDM layer can reach 70 kV mm −1 .
Aside from the amplitude of the electric field and polarity reversal, frequency and waveform are also important factors that can affect the interface charges. The recent development of power electronic devices has introduced more complex voltages that are applied to dielectric insulation systems. [11,[193][194][195] [54] Copyright 2019, IEEE.

www.advmatinterfaces.de
Hence, it is necessary to understand the effects of frequency and waveform on bulk and interface charge characteristics for insulation design and evaluation. Figure 10a shows the applied electric field waveform, which is a unipolar square wave. Figure 10b shows the space charge characteristics of epoxy resin at different frequencies with 40 kV mm −1 . [195] At low frequencies (10 Hz and 50 Hz), positive charges accumulate inside the epoxy resin and at the anode/epoxy resin interface. Slight homocharges appear at the anode interface with the increase in frequency, indicating that it is difficult for the charge to be trapped by the internal traps of the sample at high frequency. Figure 10c illustrates the sinusoidal AC voltage (amplitude 60 kV mm −1 ) waveform applied, and Figure 10d shows the effect of phase and frequency on the space charge of polyimide (PI) under sinusoidal voltage. [196] At 270°, the amount of accumulated charge decreases with the increase in frequency, which may be because the charge injection and extraction speed is very fast under high-frequency conditions, making it difficult to inject into the sample. Only a small amount of charge accumulation can be observed in the electrode/dielectric contact region. Additionally, the electric field distribution at 270° shown in Figure 10e reveals that the electric field distortion is inversely proportional to the voltage frequency. It can be seen that frequency can affect the charge accumulation and electric field distortion characteristics under different voltage conditions. The presence of charge under special voltage phases can seriously affect the dielectric properties of materials, similar to DC voltage.

Effect of Interfacial Structure
The interfacial structure of contact is a significant factor in charge behavior at the interface. [40,107,197,198] During long-term operation, the electrode material and surface roughness can change the interfacial states of the electrode/dielectric/electrode structure due to defects and electrochemical corrosion. [199] Figure 11a illustrates the interface charge density and electric field at the cathode/polyethylene interfaces, [190] which reveals that the interface charge density reaches a maximum with time for both semiconductor and metal electrode structures. Additionally, the interface charge accumulation at the Au electrode contact is lower than that of the semiconductor and Al contact. Furthermore, the energy band structures of different electrode materials have been analyzed, with emphasis on the work function, Fermi level, and other relevant parameters. Figure 11b shows the preparation and storage of XLPE with varying surface roughness, [200] which is obtained from cable cutting orientation. Charge dynamics of XLPE with different surface roughness are presented in Figure 11c. [106] The dotted www.advmatinterfaces.de line region shows that the accumulated charges at the interface increase with the roughness of the contact interfaces.
Charge injection and extraction theory suggest that the height of the injection barrier is determined by the electronic structures in metal/dielectric interfaces, including the work function of the metal, Fermi level, band gap, and electron affinity of the dielectric material. [88] As the work function of Au is larger than that of Al, the injection barrier of the Au/polymer interface is higher than that of the Al/polymer interface. [107] Consequently, more interface charge accumulation occurs at the Al/dielectric interface. In the case of a semiconductor electrode, which is commonly used in the PEA method, it is typically fabricated using carbon-doped polymer, [201] including contact resistance, defects, and voids. As a result, trapping sites can be introduced into the band structure at the electrode/dielectric interface, leading to electric field distortion and changes in charge injection. [106]

Effect of Moisture and Material Degradation
In addition to temperature, applied electric field, and electrode properties, moisture [68] and material aging [202][203][204][205] can also increase the interface charge, which can accelerate the degradation process and insulation breakdown of the dielectric system. Moisture has a significant impact on charge transport. [68,206] The water layer at the oil-impregnated paper interface introduces hydrogen bonds, leading to a reduction in trap energy, and therefore decreasing the injection barrier at the electrode interface. In the case of thermal aging of dielectric materials, thermal stress enhances charge transport. [207,208] Material deterioration due to thermal aging introduces more impurities, voids, small molecules, and byproducts, leading to an increase in deep shallow traps. [30,209,210] This contributes to the charge accumulation at the interface. The space charge behavior of thermally aged double-layered polyimide films shows that the polarity of interface charge changes from positive to negative with the increase of aging time. [203] Additionally, the interface charge density increases after aging. It can be concluded that large negative charges at the polyimide interface are caused by deep trapping levels and large trap density during thermal aging. By controlling humidity and material modification, the charge injection barrier and traps can be adjusted to improve interface charge characteristics.  [195] Copyright 2022, IEEE. (c-e) Reproduced under terms of the CC-BY license. [196] Copyright 2021, The Authors, published by Institution of Engineering and Technology. www.advmatinterfaces.de

Charge Characteristics at Electrode/Dielectric Interface
The interface scale ranges from nanometer to submicron. The charge behavior at the electrode/dielectric interface region can be obtained by detecting the surface potential using the probe microscopy method. [87,89] Figure 12a shows the surface topography of the polyetherimide (PEI)/PCBM (A kind of molecular semiconductor, ([6,6]-Phenyl C61 butyric acid methyl ester)) composite near the electrode/dielectric contact region. [211] The height of the Al electrode formed on the substrate by vacuum deposition technology is higher than that of the composite. Figure 12b,c shows the surface potential near the cathode/ composite interface. An obvious negative surface potential signal occurs along the y direction at the composite side. Since there is no negative potential signal in the interface region without charge injection (no bias voltage), the negative potential is attributed to the accumulation of injected electrons (Figure 12d). The research shows that the fillers (PCBM) in the composite immobilize free electrons via strong electrostatic adsorption and hinder the charge injection and transmission, leading to the accumulation of charges at the electrode/dielectric interface. Figure 13 shows the surface potential and charge accumulation at the electrode/SiN x dielectric interface. [89] Figure 13a illustrates the electrode/dielectric contact specimen prepared by plasma-enhanced chemical vapor deposition. A 300 nm thick SiN x layer is deposited on a low-resistivity silicon wafer, and the aluminum electrode is buried by the lift-off process. Furthermore, the SiN x passivation layer is deposited to cover the embedded electrode to prevent discharge between the electrode and the AFM tip. Figure 13b shows the surface morphology of SiN x in the test area. Maintaining a small step (<5 nm) between the dielectric and the electrode helps to avoid crosstalk between the morphology and surface potential. The white arrow indicates a topographic inhomogeneity on the surface of the sample. Figure 13c displays the surface potential curve under different bias voltages. The surface potential near the electrode/ dielectric interface increases with the increase in polarization voltage. Figure 13d shows the charge density curve calculated using Equation (10). The homogeneous charge density near the  [190] Copyright 2012, American Institute of Physics. (b) Reproduced under terms of the CC-BY license. [200] Copyright 2020, The Authors, published by Multidisciplinary Digital Publishing Institute. (c) Reproduced under terms of the CC-BY license. [106] Copyright 2020, The Authors, published by Multidisciplinary Digital Publishing Institute.

www.advmatinterfaces.de
interface increases significantly with the increase in voltage. Interestingly, the charge remains near the electrode within 180 min of polarization, resulting in a stable interface charge accumulation. The interface charge accumulation is determined by the high injection barrier (reduced by only 0.1 eV under the applied electric field). The charge carriers are difficult to overcome the injection barrier and contribute to the band conduction directly. However, the carriers can inject into the interfacial traps, leading to charge accumulation.
Although researchers have made progress in studying the charge injection mechanism at the molecular level, there are still issues in understanding the interface charges at the micro-scale. It has been demonstrated that charge injection is controlled by the electron and hole injection barrier, which is determined by the electronic properties of the metal (work function, Fermi level) and the polymer dielectric (Fermi level, band gap, electron affinity, etc.). However, since the electric field at the interface is affected by the applied voltage and accumulated charge, the injected barrier value will further change. Therefore, it is necessary to investigate the potential barrier of polymer dielectric under different conditions. The experimental scheme, combined with KPFM technology, is crucial to explain the interface charges. Although the surface potential measurement combined with Poisson's equation can calculate the charge density, the results may present large fluctuations and be affected by surface electrostatic charges, [75,77,87,89] making it challenging to study the interface charge characteristics quantitatively.

Charge Characteristics at Composite Interface
Nanocomposites exhibit better dielectric properties compared to pure polymers. [38,138,[212][213][214][215] Many researches have focused on understanding the role of the interface between the nanofiller and polymer matrix in enhancing the dielectric performance of nanocomposites. [3,216,217] Researchers have used probe microscopy technology to study the charge characteristics at the filler/ matrix interface. [218,219] Figure 14a,c show the KPFM test configurations that apply a vertical DC electric field through the AFM tip and a transverse DC electric field through the transverse electrode, respectively. Figure 14b shows the surface potential measured in the LDPE/Al 2 O 3 nanocomposite under different DC bias voltages. [162] When scanning with a negative DC bias voltage, the surface potential near the nanoparticles is significantly higher than that of LDPE. This phenomenon can be attributed to the hole injection when the probe repeatedly strikes the surface, filling the shallow traps near the particles. Similarly, when scanning with a negative DC bias voltage, electrons are injected into the interface region of the nanoparticles/ matrix, which compensated for the positive charges due to the hole injection. Figure 14d illustrates the surface potential signal of poly(vinylidene fluoride-trifluoroethylene) (P(VDF-TrFE))/ barium titanate (BT) nanocomposites under a negative electric field at 25 and 60 °C. [31] A large number of charges are trapped in the interface region at 25 °C, resulting in a strong signal near the nanoparticles. However, the halo signals near Figure 12. Charge accumulation at electrode/composite interface. a) Topography of the PEI/PCBM composite. b) Surface potential mapping of the PEI/PCBM composite. c) Surface potential profiles along the horizontal axis derived from the surface potential mappings. d) Schematic illustration of the built-in electric field formed by the molecular semiconductor captured electrons. a-d) Reproduced under terms of the CC-BY license. [211] Copyright 2020, The Authors, published by Springer Nature.
www.advmatinterfaces.de the nanoparticles (interface region) disappeared at 60 °C. This suggests that the trapped charges in the interfacial region can be released at higher temperatures. Thermally stimulated depo-larization current (TSDC) is used to study dielectric relaxation, trap energy level, transport of space charge, and molecular motion of polymer by analyzing the depolarization current The gold substrate is connected to the ground. Three nanoparticles are shown, one of which is completely embedded in the LDPE matrix. b) KPFM study of the Al 2 O 3 nanoparticles embedded in LDPE. From left to right, surface potential under the condition of none, positive, and negative DC bias, respectively. c) Schematic diagram of KPFM test configuration with DC voltage (red dotted line area) applied through the transverse electrode. d) Surface potential signals under the negative applied lateral electric field corresponding to the space charge distribution at room temperature (25 °C) and 60 °C. e) Thermally stimulated depolarization current spectra of P(VDF-TrFE)/BT nanocomposites. (a, b) Reproduced under terms of the CC-BY license. [162] Copyright 2016, The Authors, published by American Chemical Society. (c-e) Reproduced with permission. [31] Copyright 2020, American Chemical Society. www.advmatinterfaces.de generated during the sample heating process. Therefore, this phenomenon can be explained by the TSDC curve (Figure 14e). The energy level of the filler/matrix interfacial traps range from 0.71 to 0.94 eV, and the corresponding temperature range from −21 to 60 °C. When the temperature is higher than 60 °C, the trapped charge at room temperature can obtain enough energy to escape from the traps. Thus, the interfacial traps are unable to effectively capture the charges, resulting in a small potential signal observed in the interface region.

Summary and Prospect
This paper provides a review of recent research progress in interface charges at solid polymer dielectric contacts, with a focus on acoustic and probe microscopy methods. The formation of interface charge at dielectric contacts is determined by the typical MWS polarization and interface electronic structure. However, quantitative analysis of interface charge is complicated due to the limitations of current characterization and calculation techniques. MWS polarization is useful for interface charge analysis in multi-layered structures that involve different materials on a large scale. Meanwhile, electronic structure, including trapping sites and energy levels of impurities, can be used to discuss interface charges on a micro-scale, especially in small electronic devices. Acoustic methods, such as the PEA method, have indicated that interface charge accumulation and decay depend on temperature and electric field. However, the low spatial resolution in thin films due to the response of the piezoelectric sensor and pressure wave is a limitation. Fortunately, the development of in situ measurement techniques now provides nanoscale charge mapping of electrical properties. For instance, KPFM offers a powerful tool for analyzing the interface charge between the nanofiller and polymer matrix at the nanoscale. Interestingly, incorporating nanoparticles into the polymer matrix can increase the trap density at the interface region, which improves charge accumulation.
Acoustic and probe microscopy methods are essential for studying interface charges in polymer dielectric contacts. However, several key issues still need to be addressed in the future. First, thick samples (≈mm) experience significant acoustic attenuation and dispersion, which makes it challenging to improve the acoustic method for measuring interface charges. Second, the thickness of the piezoelectric sensor and pulse width of the pulse source limits the resolution of the acoustic method for measuring interface charges in thin film samples (≈µm). Third, measuring charge injection and extraction at the interface under high-frequency non-sinusoidal voltages is still challenging for both acoustic and KPFM methods. Fourth, eliminating the interference of the surface electrostatic charge of the sample remains an issue. Topography scanning in tap mode may lead to contact charges between the probe and the sample, which interferes with the interface charge measurement. Fifth, interface charge extraction from the potential signal and probe calibration still needs improvement. Lastly, the preparation of the sample surface has been neglected. It is necessary to ensure that the sample surface is flat and thin enough to reflect the internal charge transport characteristics of the sample through the interface charge. Addressing these challenges will help advance the understanding of interface charges in polymer dielectric contacts.
Further research on interface charge measurement and analysis is necessary to understand the characteristics of interface charges in polymer dielectric contacts. In the case of the acoustic method, it is essential to quantitatively discuss the positive and negative charges, particularly regarding the charge origin. Moreover, signal processing and calibration of interface charges are required based on the MWS polarization and charge injection. High-resolution PEA methods with a narrow pulse width (0.5 ns) and very thin piezoelectric sensor film (<1 µm), as well as techniques based on femtosecond lasers, can characterize interface charges in power electronic and micro-electromechanical systems fields. In thick samples, test results can be calibrated by compensating for the attenuation and dispersion of sound waves. High-frequency voltage applications are dominant factors in dielectric degradation and breakdown. Therefore, it is urgent to study interface charge measurement induced by a high voltage change ratio (high dv/ dt), requiring both high-resolution and paid responses in this high-frequency application. Regarding microscopy techniques, the main focus of research should be on the development of calibration methods to eliminate the crosstalk of surface static charges, contact electrification, and sample morphology on the measurement of interface charges, expand the application scenarios of KPFM, and adopt novel technologies to improve the sensitivity and resolution of KPFM. Furthermore, the recent Terahertz (THz) technique can enhance the resolution of interface charge measurement and should be studied further, considering the stability of the THz excitation system and following the nanoscale signal processing method. Understanding interface charges using advanced acoustic, optical, and electric techniques could drive the development of polymer dielectric design and operation in high-power electronic and microelectronic devices.