Fe/Mn Multilayer Nanowires as High-Performance T1-T2 Dual Modal MRI Contrast Agents

A lot of nanomaterials are using T1-T2 dual mode magnetic resonance (MR) contrast agents (CAs), but multilayer nanowire (NW) with iron (Fe) and manganese (Mn) as T1-T2 dual modal CAs has not been reported yet. Herein, we synthesized a Fe/Mn multilayer NW with an adjustable Fe layer, as T1-T2 dual-mode CAs. The relaxation performance of Fe/Mn multilayer NW was studied at 1.5 T. Results show that, when the length of the Fe layer is about 10 nm and the Mn is about 5 nm, the r1 value (21.8 mM−1s−1) and r2 value (74.8 mM−1s−1) of the Fe/Mn multilayer NW are higher than that of Mn NW (3.7 mM−1s−1) and Fe NW (59.3 mM−1s−1), respectively. We predict that our Fe/Mn multilayer NW could be used as T1-T2 dual mode MRI CAs in the near future.


Introduction
Magnetic resonance imaging (MRI) is a very powerful technique, widely used in clinical medical imaging technology and in scientific research. It uses static magnetic fields and radiofrequency magnetic fields to image human tissues. In the imaging process, neither ionization radiation nor a contrast agent can be used to obtain clear images with high contrast. The application of MRI in medical diagnosis can be roughly summarized into seven aspects: (1) tumor diagnosis and evaluation [1], (2) vascular display [2], (3) dynamic joint observation and detection of small lesions [3], (4) diagnosis and tracking of stroke [4], (5) cardiac imaging and cardiac function assessment, (6) surgery and interventional therapy [5].
In general, the use of CAs can enhance contrast between the area of interest (RIO) and the surrounding tissues, thus improving the sensitivity of the MRI [6,7], and can interact with the surrounding water particles to shorten their relaxation time [8][9][10], thus producing brighter and darker images on T 1 and T 2 images [11][12][13]. After years of development, CAs are broadly divided into T 1 CAs and T 2 CAs. T 1 CAs are usually paramagnetic, including gadolinium complexes, manganese salts or manganese chelates, and the Gd 3 -based complexes [14][15][16]. Manganese chelates (new dendritic DTPA manganese chelates [17], Mn-DOTA [18]) are mostly used in clinical diagnosis nowadays. T 2 CAs are usually superparamagnetic particles and ferromagnetic particles, and the superparamagnetic iron oxide nanoparticles (SPIOS) are the most widely studied and can also be used in hyperthermia [19], in addition to being often used as T 2 CAs [20,21]. Although the development of CAs has achieved great success, this single-mode contrast agent is facing many challenges in the new technology. Positive CAs generally have large permanent magnetic moments, can shorten T 1 relaxation time, improving imaging sensitivity, but they are more toxic and take less time to circulate [22,23]. Negative CAs, despite their large saturation magnetization, can help distinguish between normal and diseased tissues and have a long from Bayer Healthcare Co., LTD. (Beijing, China). TEOS and (NH 4 ) 2 SO 4 were purchased from Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). All chemicals were used as received without further purification.
The surface structure of an aluminum anodic oxide (AAO) template with NWs was observed using a field emission scanning electron microscope (FESEM, ZEISS-Utral55, Germany). The TEM (transmission electron microscopy, JEOL-2100F, Japan) was used to study the structure of NWs. In order to release the NWs from the AAO template for TEM observation, first they were dissolved in a 5% NaOH solution, then washed with DI water many times. Finally, the NWs were dispersed in the ethanol solution. X-ray diffraction (XRD, Germany) was used to analyze the crystal structure of the NWs. To study the distribution data of the elements, EDX was done on JEOL ARM-200F FESEM (Japan), operating at 200 kV accelerating voltage. An energy dispersive spectrometer (EDS, ZEISS-Utral55) was used to study the elemental composition of NWs. The SiO 2 coated on the surface of NWs was characterized by IR Prestige-21. The concentration of elements in the nanowire dispersion system was measured by Optima 7300 DV. The zeta potential of the nanowire dispersion system was characterized by Zetasizer (UK). The magnetic properties of NWs were analyzed by ADE Model4HF VSM under a magnetic field up to 20kOe. The T 1 or T 2 phantom MR images were obtained on a 1.5 T scanner (Symphony, Siemens Medical systems, Erlangen, Germany).

Synthesis of Fe and Mn NWs
The synthesis of an AAO template was discussed in our previously published paper [36], this process uses a two-step anodization method. AAO templates with 30 nm diameter were used to synthesize the Fe NWs. The electrochemical deposition of Fe NWs was done using an electrolytic solution of 80.4 g/L FeSO 4 ·7H 2 O, 25g/L H 3 BO 3 , and 0.1 g/L ascorbic acid at −1.5 V for 36 s using the direct current deposition technique. AAO templates with 30 nm diameter were also used to synthesize the Mn NWs using electrodeposition, using direct current deposition at −2.0 V for 65 s. The composition of electrolytes was 120 g/L MnSO 4 , 130 g/L (NH 4 ) 2 SO 4, and 1 g/LSnO 2 . The pH of the solution was adjusted to 6.0-7.5 by adding a little ammonia and sulfuric acid solution.
The three-electrode cell was used during the electrochemical deposition at room temperature. The cathode was the AAO templates, which sputtered gold layer at the back side, the anode was platinum electrode, and the reference electrode was a saturated calomel electrode.

Synthesis of Fe/Mn Multilayer NWs
For the electrochemical deposition of Fe/Mn multilayer NWs, the composition of the electrolyte used was 1.5 g/L FeSO 4 ·7H 2 O, 215 g/L MnSO 4 , 130 g/L (NH 4 ) 2 SO 4 , 1 g/L SnO 2 , and 25 g/L H 3 BO 3 . The solution pH was adjusted to 6.0-7.5 by adding a little H 2 SO 4 and NH 3 solution. The deposition voltage, deposition time, and cyclic deposition times of Fe and Mn layers are shown in Table 1. The template was washed with deionized water and dried after deposition. The deposition process is shown in Figure 1.

Surface Functionalization of the NWs
Because the magnetic NWs have poor hydrophilicity, they can easily gather; hence, they can not be used directly to analyze the relaxation efficiency. So the Fe/Mn multilayer NWs were functionalized with mSiO2, as silica has good biocompatibility, is known to be nontoxic, harmless and hydrophilic, and has chemical, mechanical, and thermal stability [7,21,34]. The specific operation is as follows: A certain number of prepared NWs were evenly dispersed in the mixed solution of 140 mL of ethanol and 20 mL of distilled water to form a colloidal solution. After ultrasonic treatment for 30 min, the NWs were fully dispersed in the solution. Then, 2 mL of concentrated ammonia water was added into the solution. After 30 min of mechanical stirring, a mixed solution of 1.5 mL TEOS and 10 mL ethanol were added drop-wise into the colloidal solution and stirred for 3 h at room temperature. In order to remove the excess SiO2 impurities and the residual ammonia water, after the reaction, the solution was washed repeatedly with ethanol and distilled water until it was no longer turbid. The precipitate was dried in a vacuum drying oven for standby use. We could also coat the Fe NWs and Mn NWs with mSiO2 using the above method.

MRI in Solution
The Fe/Mn@SiO2 multilayer NWs sample was configured with a dispersed solution with a concentration gradient of Mn (0~0.5 mm) and a concentration gradient of Fe (0~0.5 mm), and whose longitudinal relaxation rates, 1/T1, and transverse relaxation rates, 1/T2, were measured respectively. Their relaxation degrees, r1 and r2, were obtained through the slope of the fitting line. The

Surface Functionalization of the NWs
Because the magnetic NWs have poor hydrophilicity, they can easily gather; hence, they can not be used directly to analyze the relaxation efficiency. So the Fe/Mn multilayer NWs were functionalized with mSiO 2 , as silica has good biocompatibility, is known to be nontoxic, harmless and hydrophilic, and has chemical, mechanical, and thermal stability [7,21,34]. The specific operation is as follows: A certain number of prepared NWs were evenly dispersed in the mixed solution of 140 mL of ethanol and 20 mL of distilled water to form a colloidal solution. After ultrasonic treatment for 30 min, the NWs were fully dispersed in the solution. Then, 2 mL of concentrated ammonia water was added into the solution. After 30 min of mechanical stirring, a mixed solution of 1.5 mL TEOS and 10 mL ethanol were added drop-wise into the colloidal solution and stirred for 3 h at room temperature. In order to remove the excess SiO 2 impurities and the residual ammonia water, after the reaction, the solution was washed repeatedly with ethanol and distilled water until it was no longer turbid. The precipitate was dried in a vacuum drying oven for standby use. We could also coat the Fe NWs and Mn NWs with mSiO 2 using the above method.

MRI in Solution
The Fe/Mn@SiO 2 multilayer NWs sample was configured with a dispersed solution with a concentration gradient of Mn (0~0.5 mm) and a concentration gradient of Fe (0~0.5 mm), and whose longitudinal relaxation rates, 1/T 1 , and transverse relaxation rates, 1/T 2 , were measured respectively. Their relaxation degrees, r 1 and r 2 , were obtained through the slope of the fitting line. The

Synthesis and Characterization
Figure 2a is a FESEM diagram of the front side of the AAO template prepared in this paper. It can be seen that the template holes are uniformly distributed and orderly, and that the pore diameter is about 30 nm. Figure 2b shows the TEM images of Fe NWs; it is clear from the image that the outer wall of the as-prepared NWs was smooth and that the wire diameter was uniform. In order to accurately measure the length of the NW, we did not use ultrasonic dispersion. The measured NW's length distribution was relatively narrow. It is discernible from the inset of Figure 2b that the length of NWs was about 0.5 µm (0.521 ± 0.041 µm), and the diameter was about 30 nm (31.27 nm). A TEM diagram of Mn NWs is shown in Figure 2c. It can be seen that the morphology of Mn NWs is similar to that of Fe NWs. The length and diameter of Mn NWs was approximately 0.5 µm (0.494 ± 0.109 µm) and 30 nm (28.98 nm), respectively. Figure 2d shows the X-ray diffraction pattern of Fe NWs and Mn NWs. By comparing the diffraction peak in the figure with the Fe standard card (PDF No.06-0696) and Mn standard card (PDF No.32-0637), it can be read out that both Fe NWs and Mn NWs prepared are body-centered cubic structures.

Synthesis and Characterization
Figure 2a is a FESEM diagram of the front side of the AAO template prepared in this paper. It can be seen that the template holes are uniformly distributed and orderly, and that the pore diameter is about 30 nm. Figure 2b shows the TEM images of Fe NWs; it is clear from the image that the outer wall of the as-prepared NWs was smooth and that the wire diameter was uniform. In order to accurately measure the length of the NW, we did not use ultrasonic dispersion. The measured NW's length distribution was relatively narrow. It is discernible from the inset of Figure 2b that the length of NWs was about 0.5 μm (0.521 ± 0.041 μm), and the diameter was about 30 nm (31.27 nm). A TEM diagram of Mn NWs is shown in Figure 2c. It can be seen that the morphology of Mn NWs is similar to that of Fe NWs. The length and diameter of Mn NWs was approximately 0.5 μm (0.494 ± 0.109 μm) and 30 nm (28.98 nm), respectively. Figure 2d shows the X-ray diffraction pattern of Fe NWs and Mn NWs. By comparing the diffraction peak in the figure with the Fe standard card (PDF No.06-0696) and Mn standard card (PDF No.32-0637), it can be read out that both Fe NWs and Mn NWs prepared are body-centered cubic structures.  Since the morphology, structure, and composition of the prepared Fe/Mn multilayer NWs are similar, we choose 4# Fe/Mn multilayer NWs for characterization. Figure 3a shows the TEM image of multiple 4# Fe/Mn multilayer NWs, and the inset shows the size distribution. It can be seen that the Fe layer alternates with the Mn layer; the interface is clearly visible; the surface of the NW is smooth and the thickness is uniform. The length of the Fe/Mn multilayer NWs exhibited a narrow size distribution. The Fe/Mn NWs obtained alternating deposition 11 times and owned an average length of about 0.5 µm (0.47 ± 0.095 µm).
Since the morphology, structure, and composition of the prepared Fe/Mn multilayer NWs are similar, we choose 4# Fe/Mn multilayer NWs for characterization. Figure 3a shows the TEM image of multiple 4# Fe/Mn multilayer NWs, and the inset shows the size distribution. It can be seen that the Fe layer alternates with the Mn layer; the interface is clearly visible; the surface of the NW is smooth and the thickness is uniform. The length of the Fe/Mn multilayer NWs exhibited a narrow size distribution. The Fe/Mn NWs obtained alternating deposition 11 times and owned an average length of about 0.5 μm (0.47 ± 0.095 μm).  The appearance of the weak Au (111) diffraction peak is mainly caused by the gold plating layer on the back of the alumina template. In addition, the existence of Fe3Mn7 indicates that codeposition of two metals occurs during the Fe, Mn alternating deposition.
A magnified individual Fe/Mn multilayer NW shown in Figure 3c indicates the diameter was about 30 nm (30.8 nm). The corresponding SAED pattern shown in the upper-right inset in Figure 3c revealed the polycrystalline structure of the Fe/Mn NWs.
Owing to the fact that the backscattering ability of Fe is greater than that of Mn, and the lining of the Fe layer is higher than the Mn layer, the darker part should be Mn with a length of about 5 nm (4.6 nm), while the brighter part corresponds to Fe with a length of about 40 nm (38.6 nm). The illustration is the electron diffraction pattern of the nanorod, indicating that both Fe and Mn layers are polycrystalline.
In order to further confirm the composition and the distribution of the elemental metal, EDX line scanning analysis was conducted on the dotted line shown in TEM Figure 3c. The result is shown in Figure 3d; the solid line represents the Fe components and the dotted line represents the Mn component. The diffraction peaks corresponding to the two components show continuous alternating on the whole, and the result was consistent with the multilayer structure observed in the TEM image in Figure 3c. Moreover, we find that, in the energy spectrum of the Mn layer, in addition to the characteristic peaks of the Mn elements, there are weak peaks of the Fe elements, but in the energy spectrum of the Fe layer, there is no corresponding characteristic peak of the Mn elements. The results indicate that the Fe layer in Fe/Mn multilayer NWs is mainly composed of Fe phase with bcc structure, while the Mn layer is composed of simple elements and the alloy Fe3Mn7 with bcc structure. The phase Fe3Mn7 occurs with the deposition of Mn components during the formation of the layer. In order to investigate the element composition of Fe/Mn multilayered NWs, we carried out an energy spectrum analysis on the same NW. The results are shown in Figure 3e. The atomic ratio of Fe to Mn is about 9.4:1. Figure 3f shows the measured metal element composition of the multiple NWs in Figure 3a by surface scanning, and the atomic ratio of Fe and Mn is about 8.7:1. The atomic ratio of Fe and Mn, measured twice, is very close, so the single Fe/Mn multilayered nanowires are representative. In addition to Fe and Mn, there are also O, Al, Cu, Na, and C in the EDS spectra. Among them, O and Al are the components of the alumina template. C comes from the C layer sprayed before the sample test, and Cu and Na come from the copper mesh and NaOH respectively.
The position of the Mn layer in Fe/Mn multilayer NWs was analyzed by energy spectrum and the results are shown in Table 2. The atomic percentage of Fe and Mn in the Mn layer is about 12.59:87.41, which means that the Mn layer is actually a Mn0.87Fe0.13 alloy layer. Although the concentration of Fe in the bath is very low, the deposition of Fe is accompanied by the deposition of the Mn layer. It is very important to control the concentration ratio of Fe and Mn in order to ensure the deposition of Fe layer and avoid the codeposition of too much Fe in the Mn layer.  Figure 4 is the TEM diagram of 1#~4# Fe/Mn multilayer NWs prepared using different Fe deposition times. We can see that the four NWs are of uniform thickness and have a smooth surface, and the interface between the Mn layer and the Fe layer is clearly visible; the length of each layer is uniform. The length and diameter of the four NWs are listed in Table 3. It can be seen that the diameter (about 30 nm) and the length (about 0.5 µm) of the four NWs are relatively similar. With the increase in Fe deposition time, the length of the Fe layer increases correspondingly. The length of the Mn layer in the 1#~4# Fe/Mn multilayer NWs is about 5 nm, while the length of the Fe layer is approximately 5, 10, 20, and 40 nm, respectively. The growth rate of the Fe layer increases gradually from 2 to 3.3 nm/s, and the growth rate of the Mn layer is about 5 nm. The results show that the length of each layer can be adjusted by controlling the deposition time of each layer of the Fe/Mn multilayer NWs. Therefore, we can obtain multilayer NWs with different diameters and sublayer lengths using electrodeposition. a smooth surface, and the interface between the Mn layer and the Fe layer is clearly visible; the length of each layer is uniform. The length and diameter of the four NWs are listed in Table 3. It can be seen that the diameter (about 30 nm) and the length (about 0.5 μm) of the four NWs are relatively similar. With the increase in Fe deposition time, the length of the Fe layer increases correspondingly. The length of the Mn layer in the 1#~4# Fe/Mn multilayer NWs is about 5 nm, while the length of the Fe layer is approximately 5, 10, 20, and 40 nm, respectively. The growth rate of the Fe layer increases gradually from 2 to 3.3 nm/s, and the growth rate of the Mn layer is about 5 nm. The results show that the length of each layer can be adjusted by controlling the deposition time of each layer of the Fe/Mn multilayer NWs. Therefore, we can obtain multilayer NWs with different diameters and sublayer lengths using electrodeposition.  The length of Mn in 1# Fe/Mn multilayer NWs looks smaller than in 4# Fe/Mn multilayer NWs from TEM images in Figure 4. This may be because the Mn layer at the initial sedimentary stage is actually the competition of the Fe 2+ and Mn 2+ sedimentary processes before the start of the sedimentary Mn layers. 4# Fe/Mn multilayer NWs, due to the longer  The length of Mn in 1# Fe/Mn multilayer NWs looks smaller than in 4# Fe/Mn multilayer NWs from TEM images in Figure 4. This may be because the Mn layer at the initial sedimentary stage is actually the competition of the Fe 2+ and Mn 2+ sedimentary processes before the start of the sedimentary Mn layers. 4# Fe/Mn multilayer NWs, due to the longer Fe deposits, result lower Fe 2+ concentration in channel, compared with 1# Fe/Mn multilayer NWs, so in the initial stages of the sedimentary Mn layer 1# Fe/Mn multilayer NWs, accompanied by more codeposition of Fe 2 + , leads to Mn layer thickness looking shorter than that of 4# Fe/Mn multilayer NWs.

Water-Solubilization of NWs
Because the morphology and composition of NWs in this paper are similar, they have little influence on the SiO 2 coating process, so we selected the 4# Fe/Mn multilayer NWs covered by SiO 2 for characterization. A typical TEM image of the as-prepared Fe/Mn@SiO 2 NWs is shown in Figure 5a. It can be seen from the figure that, after coating with SiO 2 , the NWs are relatively dispersed without a large amount of aggregation. An enlarged TEM image of an individual Fe/Mn@SiO 2 NW in the inset indicates the core-shell structured NW has a uniform diameter of about 90 nm (86.7 nm) and a mesoprous silica layer with a thickness of 30 nm. It can be seen that the central part of each NW is darker than the outer layer, due to the fact that the atomic number of Fe or Mn is higher than that of Si, which confirms the core-shell structure of the NWs. A FT-IR spectrum analysis was performed on Fe/Mn@SiO 2 multilayer NWs, and the results are shown in Figure 5b. The measured spectrum was completely consistent with the typical characteristic absorption bands of amorphous SiO 2 , indicating that our NWs were successfully covered by SiO 2 [37].

Water-Solubilization of NWs
Because the morphology and composition of NWs in this paper are similar, they have little influence on the SiO2 coating process, so we selected the 4# Fe/Mn multilayer NWs covered by SiO2 for characterization. A typical TEM image of the as-prepared Fe/Mn@SiO2 NWs is shown in Figure 5a. It can be seen from the figure that, after coating with SiO2, the NWs are relatively dispersed without a large amount of aggregation. An enlarged TEM image of an individual Fe/Mn@SiO2 NW in the inset indicates the core-shell structured NW has a uniform diameter of about 90 nm (86.7 nm) and a mesoprous silica layer with a thickness of 30 nm. It can be seen that the central part of each NW is darker than the outer layer, due to the fact that the atomic number of Fe or Mn is higher than that of Si, which confirms the core-shell structure of the NWs. A FT-IR spectrum analysis was performed on Fe/Mn@SiO2 multilayer NWs, and the results are shown in Figure 5b. The measured spectrum was completely consistent with the typical characteristic absorption bands of amorphous SiO2, indicating that our NWs were successfully covered by SiO2 [37]. The Fe/Mn multilayer NWs coated with SiO2 were installed into the centrifugal tube and dispersed with water. A dispersion solution with Fe concentration of about 0.521 mM was prepared. Figure 5c is the zeta potential diagram of the sample prepared. The results show that the Fe/Mn multilayer NW's dispersion has a high zeta potential (−42.9 mV), indicating that the NWs coated with SiO2 have good dispersion.

Analysis of Magnetic Behavior
The magnetic properties of 1#~4# Fe/Mn@SiO2 multilayer NWs and Fe NWs were measured by VSM. The results are shown in Figure 6. The magnetic properties calculated from the hysteresis loop are listed in Table 4. The Fe/Mn multilayer NWs coated with SiO 2 were installed into the centrifugal tube and dispersed with water. A dispersion solution with Fe concentration of about 0.521 mM was prepared. Figure 5c is the zeta potential diagram of the sample prepared. The results show that the Fe/Mn multilayer NW's dispersion has a high zeta potential (−42.9 mV), indicating that the NWs coated with SiO 2 have good dispersion.

Analysis of Magnetic Behavior
The magnetic properties of 1#~4# Fe/Mn@SiO 2 multilayer NWs and Fe NWs were measured by VSM. The results are shown in Figure 6. The magnetic properties calculated from the hysteresis loop are listed in Table 4.  We can see in Figure 6a, as for Fe/Mn@SiO2 multilayer NWs, the length of the Fe layer  We can see in Figure 6a, as for Fe/Mn@SiO 2 multilayer NWs, the length of the Fe layer (5 nm-40 nm) is proportional to the Hc (8.2 Oe~81.4 Oe) and Mr (0.1 emu/g~12.2 emu/g) of the Fe/Mn multilayer NWs. This may be because the Fe layer is in a single domain structure with a relatively small length, and the Hc and Mr are proportional to the size. When the length of the Fe layer is small to a certain extent, Fe/Mn multilayer NWs almost disappear and NWs show superparamagnetic characteristics, which lay a foundation for Fe/Mn multilayer NWs to become a magnetic resonance T 1 -T 2 dual-mode contrast agent. In addition, we found that, with the decrease in the length of the Fe layer, the saturation magnetization of the Fe/Mn multilayer NWs decreased gradually (43.2~26.5 emu/g) and finally decreased significantly (11.3 emu/g), which was mainly caused by the decrease in the length of the Fe layer and the decrease of the Fe component relative to the Fe/Mn multilayer NWs. When the length of the Fe layer decreases to 5 nm, the size effect of the Fe layer becomes more obvious. At this point, the atomic structure and symmetry of the Fe layer are different from those of the internal atoms, so the saturation magnetization will be strongly reduced. As can be seen from the hysteresis cycle of Fe NWs in Figure 6b, Fe NWs exhibit ferromagnetism, and their saturation magnetization (48.1 emu/g) and coercive force (658 Oe) are significantly greater than Fe/Mn multilayer NWs, which is mainly due to the magnetic anisotropy caused by the shape anisotropy of Fe NWs.

The Relaxation Efficiency of Fe/Mn Multilayered NWs
To validate the potential of the Fe/Mn@SiO 2 multilayer NWs as T 1 -T 2 dual modal CAs, we performed their phantom MR imaging on a 1.5 T MRI scanner and calculated both their longitudinal and transverse relaxivities r 1 and r 2 , respectively. Using self-prepared Fe NWs and Mn NWs, which were similar in size to multilayer NWs, as respective T 1 and T 2 controls, samples of the Fe/Mn@SiO 2 multilayer NWs containing different metal concentrations were scanned. Figure 7a,b show T 1 and T 2 relaxation properties of Fe/Mn@SiO 2 multilayer NWs measured respectively, and the relevant data are listed in Table 5.  We can see in Figure 6a, as for Fe/Mn@SiO2 multilayer NWs, the length of the Fe layer (5 nm-40 nm) is proportional to the Hc (8.2 Oe~81.4 Oe) and Mr (0.1 emu/g~12.2 emu/g) of the Fe/Mn multilayer NWs. This may be because the Fe layer is in a single domain structure with a relatively small length, and the Hc and Mr are proportional to the size. When the length of the Fe layer is small to a certain extent, Fe/Mn multilayer NWs almost disappear and NWs show superparamagnetic characteristics, which lay a foundation for Fe/Mn multilayer NWs to become a magnetic resonance T1-T2 dual-mode contrast agent. In addition, we found that, with the decrease in the length of the Fe layer, the saturation magnetization of the Fe/Mn multilayer NWs decreased gradually (43.2~26.5 emu/g) and finally decreased significantly (11.3 emu/g), which was mainly caused by the decrease in the length of the Fe layer and the decrease of the Fe component relative to the Fe/Mn multilayer NWs. When the length of the Fe layer decreases to 5 nm, the size effect of the Fe layer becomes more obvious. At this point, the atomic structure and symmetry of the Fe layer are different from those of the internal atoms, so the saturation magnetization will be strongly reduced. As can be seen from the hysteresis cycle of Fe NWs in Figure 6b, Fe NWs exhibit ferromagnetism, and their saturation magnetization (48.1 emu/g) and coercive force (658 Oe) are significantly greater than Fe/Mn multilayer NWs, which is mainly due to the magnetic anisotropy caused by the shape anisotropy of Fe NWs.

The Relaxation Efficiency of Fe/Mn Multilayered NWs
To validate the potential of the Fe/Mn@SiO2 multilayer NWs as T1-T2 dual modal CAs, we performed their phantom MR imaging on a 1.5 T MRI scanner and calculated both their longitudinal and transverse relaxivities r1 and r2, respectively. Using self-prepared Fe NWs and Mn NWs, which were similar in size to multilayer NWs, as respective T1 and T2 controls, samples of the Fe/Mn@SiO2 multilayer NWs containing different metal concentrations were scanned. Figure 7a,b show T1 and T2 relaxation properties of Fe/Mn@SiO2 multilayer NWs measured respectively, and the relevant data are listed in Table 5.  We found that the r1 (21.5 mM −1 s −1~2 4.7 mM −1 s −1 ) of 1#~4# Fe/Mn@SiO2 multilayer NWs was significantly higher than the r1 (5.7 mM −1 s −1 ) of single-component Mn@SiO2 NWs with similar morphology and size, and the r2 (74.8 mM −1 s −1 ) of 2# Fe/Mn@SiO2 multilayer NWs was higher than the r2 (59.5 mM −1 s −1 ) of single-component Fe@SiO2 NWs with similar morphology and size. r1 and r2 of 2# Fe/Mn@SiO2 multilayer NWs were higher than r1 (3.8 mM −1 s −1 ) and r2 (53 mM −1 s −1 ) of Combidex, a T1 contrast agent commonly used in clinical practice [38]. In addition, the r2/r1 values of 2# Fe/Mn@SiO2 multilayer NWs satisfy the conditions (2 < r2/r1 < 10), indicating that 2# Fe/Mn@SiO2 multilayer NWs have significant T1-T2 dual-mode contrast enhancement effects. When the length of the Fe layer is 10 nm, the r2 (74.8 mM −1 s −1 ) of the Fe/Mn@SiO2 multilayer NWs is the maximum, while, when the length of Fe layer is 5 nm, the r1 (24.7 mM −1 s −1 ) of Fe/Mn@SiO2 multilayer NWs is the maximum. Figure 7c,d show the T1and T2-weighted MRI images of Fe/Mn@SiO2 multilayer NWs under different Mn concentration gradients and different Fe concentration gradients, respectively. As can be seen from Figure 7c, with an increase in Mn concentration, the T1-weighted MRI images of Fe/Mn@SiO2 multilayer NWs become brighter and brighter. Moreover, under the same Mn concentration, Fe/Mn@SiO2 multilayer NWs have a stronger contrast than Mn@SiO2 NWs. The contrast ratio of 1# Fe/Mn@SiO2 multilayer NWs is the highest, while that of 4# Fe/Mn@SiO2 multilayer NWs is the weakest. As can be seen from Figure 7d, with an increase in Fe concentration, the T2-weighted MRI image of Fe/Mn@SiO2 multilayer NWs becomes darker and darker. Moreover, at the same Fe concentration, the contrast ratio of 2# Fe/Mn@SiO2 multilayer NWs is stronger than that  We found that the r 1 (21.5 mM −1 s −1~2 4.7 mM −1 s −1 ) of 1#~4# Fe/Mn@SiO 2 multilayer NWs was significantly higher than the r 1 (5.7 mM −1 s −1 ) of single-component Mn@SiO 2 NWs with similar morphology and size, and the r 2 (74.8 mM −1 s −1 ) of 2# Fe/Mn@SiO 2 multilayer NWs was higher than the r 2 (59.5 mM −1 s −1 ) of single-component Fe@SiO 2 NWs with similar morphology and size. r 1 and r 2 of 2# Fe/Mn@SiO 2 multilayer NWs were higher than r 1 (3.8 mM −1 s −1 ) and r 2 (53 mM −1 s −1 ) of Combidex, a T 1 contrast agent commonly used in clinical practice [38]. In addition, the r 2 /r 1 values of 2# Fe/Mn@SiO 2 multilayer NWs satisfy the conditions (2 < r 2 /r 1 < 10), indicating that 2# Fe/Mn@SiO 2 multilayer NWs have significant T 1 -T 2 dual-mode contrast enhancement effects. When the length of the Fe layer is 10 nm, the r 2 (74.8 mM −1 s −1 ) of the Fe/Mn@SiO 2 multilayer NWs is the maximum, while, when the length of Fe layer is 5 nm, the r 1 (24.7 mM −1 s −1 ) of Fe/Mn@SiO 2 multilayer NWs is the maximum. Figure 7c,d show the T 1 -and T 2 -weighted MRI images of Fe/Mn@SiO 2 multilayer NWs under different Mn concentration gradients and different Fe concentration gradients, respectively. As can be seen from Figure 7c, with an increase in Mn concentration, the T 1 -weighted MRI images of Fe/Mn@SiO 2 multilayer NWs become brighter and brighter. Moreover, under the same Mn concentration, Fe/Mn@SiO 2 multilayer NWs have a stronger contrast than Mn@SiO 2 NWs. The contrast ratio of 1# Fe/Mn@SiO 2 multilayer NWs is the highest, while that of 4# Fe/Mn@SiO 2 multilayer NWs is the weakest. As can be seen from Figure 7d, with an increase in Fe concentration, the T 2 -weighted MRI image of Fe/Mn@SiO 2 multilayer NWs becomes darker and darker. Moreover, at the same Fe concentration, the contrast ratio of 2# Fe/Mn@SiO 2 multilayer NWs is stronger than that of Fe@SiO 2 NWs, while the contrast ratio of 1# Fe/Mn@SiO 2 multilayer NWs is the weakest.
The above results are consistent with the relaxation data, indicating that the 2# Fe/Mn@SiO 2 multilayer NWs have the potential to become a T 1 -T 2 dual-mode contrast agent.
When the length of the Fe layer is different, Fe/Mn@SiO 2 multilayer NWs show different relaxation. With the decrease in the length of the Fe layer (40 nm~5 nm), the T 2 relaxation of Fe/Mn@SiO 2 multilayer NWs first increases gradually (32.1 mM −1 s −17 4.8 mM −1 s −1 ) and then decreases rapidly (28.9 mM −1 s −1 ), which may be caused by the differences in the magnetization direction of Fe/Mn@SiO 2 multilayer NWs with different lengths of the Fe layer. In general, T 2 relaxation is closely related to such factors as the saturation magnetization of the sample and the number of water molecules covered by a nonuniform magnetic field. The stronger the saturation magnetization provided by the multilayer NWs, or the greater the number of water molecules covered by the magnetic field, the faster the hydrogen protons in the water molecules lose phase, thus accelerating and shortening the T 2 relaxation process. In the Fe/Mn@SiO 2 multilayer NWs 1~4# samples prepared by us, the smaller the length-diameter ratio of the Fe layer, the more the easy magnetization direction of the Fe layer tends to be perpendicular to the long axis direction of the NW; on the contrary, the larger the length-diameter ratio is, the easier the magnetization direction tends to be parallel to the NW direction, as shown in Figure 8. When the direction of easy magnetization is perpendicular to the long axis of NWs, the effective area covered by the magnetic field of Fe/Mn@SiO 2 multilayer NWs is the largest. Therefore, the smaller the aspect ratio of the Fe layer is, the more water molecules are covered by the magnetic field generated by Fe/Mn@SiO 2 multilayer NWs and the greater the T 2 relaxation degree of Fe/Mn@SiO 2 multilayer NWs. Although the shorter the Fe layer is the lower the saturation magnetization of the Fe/Mn@SiO 2 multilayer NWs is as well, the effective area covered by the magnetic field of the Fe/Mn multilayer NWs plays a leading role in the contribution of T 2 relaxation. Therefore, the shorter the Fe layer is, the greater the T 2 relaxation of the Fe/multilayer NWs is. When the length of the Fe layer is reduced to 5 nm, the size effect of the Fe layer is obvious. The Fe atoms in the Fe layer gather on the surface, which destroys the exchange between Fe atoms. As a result, the saturation magnetization decreases obviously, and T2 relaxation decreases sharply. At this time, although the effective area covered by the magnetic field of Fe/Mn@SiO2 multilayer NWs is the largest, it still fails to offset the significant influence brought about by the decrease in saturation magnetization, resulting in a dramatic decrease in T2 relaxation and the maximum T2 relaxation r2 of 2# Fe/Mn multilayer NWs. With the decrease in the Fe layer length (40 nm~5 nm), the T1 relaxation of the When the length of the Fe layer is reduced to 5 nm, the size effect of the Fe layer is obvious. The Fe atoms in the Fe layer gather on the surface, which destroys the exchange between Fe atoms. As a result, the saturation magnetization decreases obviously, and T 2 relaxation decreases sharply. At this time, although the effective area covered by the magnetic field of Fe/Mn@SiO 2 multilayer NWs is the largest, it still fails to offset the significant influence brought about by the decrease in saturation magnetization, resulting in a dramatic decrease in T 2 relaxation and the maximum T 2 relaxation r 2 of 2# Fe/Mn multilayer NWs. With the decrease in the Fe layer length (40 nm~5 nm), the T 1 relaxation of the Fe/Mn@SiO 2 multilayer NWs is slightly affected (21.5 mM −1 s −1~2 2.3 mM −1 s −1 ). When the length of the Fe layer decreased to 5 nm, the T 1 relaxation of the Fe/Mn@SiO 2 multilayer NWs (24.7 mM −1 s −1 ) increased slightly. This may be due to the change in magnetic field environment caused by Fe/Mn@SiO 2 multilayer NWs when the Fe layer length is greater than 5 nm, which has little effect on the precession frequency of hydrogen protons. When the length of the Fe layer is 5 nm, the local magnetic field intensity produced by Fe/Mn@SiO 2 multilayer NWs makes the precession frequency of hydrogen protons close to that of the surrounding molecules, and the resonance between them accelerates the energy release of high-energy hydrogen protons. Thus, T 1 relaxation is improved.
Fe/Mn@SiO 2 multilayer NWs exhibit better T 1 and imaging effects than Mn single component NWs. This may be due to the high magnetic sensitivity of the Fe layer, which induces the local magnetic field, so that the adjacent paramagnetic Mn layer is enhanced by the induced magnetic field generated by the surrounding Fe layer, thus accelerating the longitudinal relaxation rate of the surrounding water protons. Therefore, Fe/Mn@SiO 2 multilayer NWs can significantly shorten the longitudinal relaxation time T 1 , then enhance the positive contrast effect of the Fe/Mn@SiO 2 multilayer NWs. Because the Mn layer length is only 5 nm, the proportion of Mn atoms located on the surface or interface of the Mn layer is larger than that inside the layer, and the ratio increases with the decrease of the Mn layer size. The frequency of the symmetric electrons in Mn is close to that of the water particles. The relaxation rate of the high energy proton spin nucleus is more coordinated with that of the nearby proton spin relaxation through the formation of the electron-proton dipole interaction. The synergistic effect of these two enhancement mechanisms makes Fe/Mn@SiO 2 multilayer NWs exhibit better coverage and imaging effects. Compared with the Fe NWs, 2# Fe/Mn@SiO 2 multilayer NWs exhibit better T 2 relaxation, which may be due to 2# Fe/Mn@SiO 2 multilayer NWs providing more effective magnetic fields for hydrogen protons.
Although the saturation magnetization (32.5 emu/g) of 2# Fe/Mn@SiO 2 multilayer NWs is less than that of Fe NWs (48.3 emu/g), at this time, the increase in the effective magnetic field plays a leading role in the contribution of T 2 relaxation, which counteracts the effect of the decrease of saturation magnetization; in addition, the T 2 shortening effect from the Mn layer and the synergistic T 2 shortening effect produced by Mn near the Fe layer further promote the T 2 relaxation and imaging effects of 2# Fe/Mn@SiO 2 multilayer NWs.

a.
We have successfully developed Fe/Mn multilayer NWs with adjustable Fe layer thickness on the AAO template using the controlled potential double pulse method. The Fe layer is mainly composed of the Fe of bcc structure, and the Mn layer is composed of the Mn of bcc structure, and the Fe3Mn7 is composed of fcc structure. b.
The SiO 2 -coated Fe/Mn multilayer NWs have obvious core-shell structure. The thickness of the SiO 2 layer is uniform and has good dispersion in water. When the Fe layer length is short (≤10 nm), Fe/Mn multilayer NWs exhibit superparamagnetic characteristics. c.
MRI studies showed that, when the Fe layer length was about 10 nm and the Mn layer length was about 5 nm, the r 1 value (21.8 mM −1 s −1 ) and r 2 value (74.8 mM −1 s −1 ) were higher than the r 1 value of Mn@SiO 2 NWs (3.7 mM −1 s −1 ) and the r 2 value of Fe@SiO 2 NWs (59.3 mM −1 s −1 ). This indicates that the Fe/Mn@SiO 2 multilayer NWs show obvious T 1 positive enhancement, T 2 negative enhancement effect, and a good T 1 -T 2 dual-mode imaging effect. Data Availability Statement: Data sharing not applicable.

Conflicts of Interest:
The authors declare no conflict of interest.