-
H+ 离子束流在高速电场的作用下注入靶材中形成高阻区,该高阻区的质子浓度分布服从高斯分布函数[13]:
$$ N(x) = \frac{\varPhi }{{\sqrt {2\pi \Delta {R_P}} }}\exp \left[ { - \frac{{{{\left( {x - {R_P}} \right)}^2}}}{{2\Delta {R_P}^2}}} \right] $$ (1.1) 式中:
$ N(x) $ 为注入深度$ x $ 处的质子浓度;$\varPhi$ 为质子注入剂量;$ x $ 为距离外延层表面的深度;$ {R_P} $ 为质子入射到靶材的平均射程;$ \Delta {R_P} $ 为平均射程的标准偏差;$ 2\Delta {R_P} $ 对应注入区的结深。当注入深度$ x = {R_P} $ 时,$ {N_{\max }} $ 对应质子峰值浓度分布 。$ {R_P} $ 与注入能量$ E $ 有关,其关系式表示如下[13]:$$ R_{P}=\int_{0}^{E} \frac{{\rm{d}} E^{\prime}}{S \int \mathrm{d} \delta T \cdot \cos \theta} \exp \int_{0}^{E^{\prime}} \int {\rm{d}} \delta(1-\cos \theta) \dfrac{{\rm{d}} E^{\prime \prime}}{\int \mathrm{d} \delta T \cdot \cos \theta} $$ (1.2) 式中:
$ S $ 为靶材密度;$ \theta $ 为散射角;$ T $ 为离子入射散射损耗的能量;$ {\text{d}}\delta $ 为电子核与电子微分散射截面之和。公式(1)、(2)反映了质子浓度随注入能量和剂量的分布关系。H+离子以相同的注入剂量(6×1014 cm−2)、不同注入能量入射到GaAs材料中的平均射程与能量的关系曲线如图1所示。由图1可知,注入质子的平均射程
$ {R_P} $ 随注入能量的增加而增加,能量和平均射程比约为0.9 μm/100 keV。图 1 不同能量的H+离子注入时质子平均射程分布曲线
Figure 1. Average range distribution curve of proton during H+ ion implantation with different energy
上述注入条件下质子浓度随注入深度的函数关系曲线如图2(a)所示。由图2(a)可知,注入区质子峰值浓度随着注入能量的增大而下降。这是由于离子注入区歧变现象,注入能量越大,离子的分布将会横向分散,进而使得注入区浓度降低。
图 2 (a)不同能量的H+离子注入时质子浓度分布曲线;(b)不同剂量的H+离子注入时质子浓度分布曲线
Figure 2. (a) Proton concentration distribution curve during H+ ion implantation with different energy; (b) Proton concentration distribution curve depth during H+ ion implantation with different dose
当H+离子以一定能量(320 keV)、不同剂量注入GaAs材料中时,其浓度分布情况如图2(b)所示。从图2(b)可以看出,随着质子注入剂量的增加,质子峰值浓度显著增加。同时,质子分布结深也呈现递增趋势。由此可知,质子注入的能量和剂量会对靶材中的质子注入区的平均射程、结深、质子浓度产生综合影响,进而影响注入区的电学性能。
-
将质子以一定能量和剂量注入至GaAs材料中时,质子与靶材间发生的电子碰撞和原子核碰撞会破坏GaAs晶体结构的有序性,产生微观晶格缺陷,其中电荷型晶格缺陷会影响材料电学特性。其在GaAs材料的导带下形成深受主能级,价带上形成深施主能级,有效抑制了多数载流子的产生。降低的多子浓度随质子注入剂量的变化关系可由公式(3)表示[14]:
$$ n(\varPhi ) = n(0) - \frac{{\Delta n}}{{\Delta \varPhi }}\varPhi $$ (1.3) 式中:
$ n(0) $ 为注入前的多子浓度;$n(\varPhi )$ 为注入质子后的多子浓度;$\dfrac{{\Delta n}}{{\Delta \varPhi }}$ 为多子去除率。同时,质子注入区的电阻率受载流子浓度影响,由公式(4)表示[14]:$$ \rho {\text{ = }}\frac{{\text{1}}}{{e(n{u_n} + p{u_P})}} $$ (1.4) 式中:
$ e $ 为电子电荷量;$ n $ 和$ p $ 分别表示电子和空穴浓度;$ {u_n} $ 和$ {u_P} $ 分别表示电子和空穴的迁移率。同时,电荷型晶格缺陷作为载流子的散射中心会降低载流子的迁移率。由公式可见,多子浓度和载流子迁移率的降低导致电阻率$\; \rho $ 升高,电阻增大。由此,质子注入区成为高阻区,约束横向电流。靶材的晶格损伤程度与注入质子剂量有关,用SRIM软件模拟了不同数量的离子轰击靶材时对靶材的损伤情况,如图3所示。随着靶材中注入离子数增多,碰撞质子在靶材中将能量依次转移给其他原子,形成的级联碰撞(Cascade)使质子扩散范围扩大,产生大量复杂缺陷并聚集为较大的缺陷团[15],如图3(a)、(b)所示的缺陷状态转变。
图 3 离子碰撞产生的晶格(a)局部缺陷和(b)大量缺陷团
Figure 3. Lattice (a) local defects and (b) a large number of defect clusters produced by ion collision
通常,使用快速退火工艺可以修复损伤并激活部分对载流子运输无贡献的缺陷。但当注入质子剂量过大时,损伤区会沉积为无序区,此时难以通过退火修复晶格损伤,使得载流子迁移率降低,电阻率下降[16]。
综上可知,质子注入的能量主要影响质子注入区深度,但随着能量的提高,质子浓度呈降低趋势;质子注入剂量影响质子区分布结深和质子浓度,当剂量过大时,会对晶格缺陷产生较大影响,注入区出现多种复合缺陷类型,呈现非晶化状态,晶格难以通过退火修复,电学阻隔效果变差。在实际制备VCSEL电流限制孔径的过程中,既要保证注入区形成良好的电流阻隔,同时也要避免有源区出现点缺陷产生非辐射复合。因此,要充分考虑注入剂量和能量的相互作用和对晶格的影响机制,精确调控两参数,使得质子注入区在有源区附近形成高阻区限制电流,且其分布不能扩散进入有源区影响发光。
-
在加速电压5 kV、探针电流12 pA的扫描条件下,采用SEM观测质子注入区形貌,在样品外延片的断面可以清晰地观察到质子注入区域。将注入样品编号为1~7组样品。上述注入能量和注入剂量下质子注入平均射程
$ {R_P} $ 、结深$ {\text{2}}\Delta {R_P} $ 以及注入区下沿距有源区距离d的实验结果如表1所示。其中,质子注入平均射程影响质子区深度,质子注入区结深影响质子区浓度,注入区下沿距有源区的距离影响VCSEL发光。表 1 SEM 表征结果
Table 1. Characterization results of SEM
Group Energy/
keVDose/
cm−2Average range,
$ {R_P} $/μmJunction depth, $ {\text{2}}\Delta {R_P} $/μm Distance from
active region,
d/μm1 280 6×1014 1.55 0.35 1.28 2 320 6×1014 1.67 0.61 1.03 3 320 8×1014 1.67 0.73 0.97 4 320 1×1015 1.68 0.84 0.9 5 320 2×1015 1.70 1.07 0.77 6 360 6×1014 1.72 0.71 0.93 7 400 2×1015 2.23 1.3 0 由表1可知,当注入剂量为6×1014 cm−2时,随着注入能量的提高(280~360 keV),质子注入的平均射程
$ {R_P} $ 显示递增趋势,同时质子分布结深渐大。当注入能量为320 keV时,随着注入剂量的增加,质子分布结深随着增大,与文中理论部分分析的质子分布随注入能量和剂量变化的分布规律基本一致。第1组和第7组注入条件下的VCSEL外延片SEM扫描图如图7(a)、(b)所示。如图7(a)所示,当用280 keV,6×1014 cm−2的注入参数轰击VCSEL外延片时,质子注入区结深为0.35 μm,距有源区距离1.28 μm,此注入参数下质子停留位置距离有源区较远。如图7(b)所示,当注入能量为400 keV、注入剂量为2×1015 cm−2时,质子区结深为1.3 μm。通过观测,注入区已经渗透到有源区,这会对器件性能造成损伤,影响出光。因此,较大的注入能量和剂量参数并不适用于VCSEL电流限制孔径的制备。
图 7 注入条件为(a) 280 keV,6×1014 cm−2; (b) 400 keV,2×1015 cm−2; (c) 320 keV,8×1014 cm−2和(d) 360 keV,6×1014 cm−2时的SEM扫描图像
Figure 7. SEM images under the injection conditions of (a) 280 keV,6×1014 cm−2; (b) 400 keV,2×1015 cm−2; (c) 320 keV,8×1014 cm−2 and (d) 360 keV,6×1014 cm−2
当注入参数为320 keV,8×1014cm−2和360 keV,6×1014 cm−2时,对应SEM扫描图像如图7(c)、(d)所示,注入区形成的结深均约0.7 μm,与有源区距离均约0.9 μm。从这两组的观测结果可直观地反映注入能量和剂量对注入效果的综合影响。第3组样品表现了以较小的能量虽轰击质子的深度较浅,但以较大的剂量注入时,依然能形成与第6组样品相似的注入结果。对于VCSEL的制备,这两者的质子分布区均满足条件,但仍需考虑质子注入区阻值。
其他以320 keV的能量注入的3组样品,其中6×1014 cm−2的注入剂量所形成的质子结深较浅,无法很好地约束横向注入电流的路径,而1015 cm−2量级的注入剂量形成的质子区分布与有源区靠近,影响器件性能。
-
观测完成后使用探针台对样品进行I-V测试,并求得电阻。样品的电阻值随注入剂量的变化曲线如图8所示。由图8可以看出,注入剂量在6×1014~8×1014 cm−2时,由于注入后引入深能级状态,复合了GaAs多数载流子[13],样品电阻值达到107~108 Ω∙cm2,可以实现优异的电流隔离效果。然而,随着注入剂量继续增大,靶材产生的空穴与间隙原子增多,损伤区不断扩大,注入层的缺陷团相互交叠构成非晶层,出现无序的状态。通过退火难以激活杂质并修复大部分晶格损伤,因此电阻值呈下降趋势[17]。
根据SEM观测和I-V测试结果,通过分析质子注入量的相互作用和影响机制,在综合考虑质子分布位置和电流隔离效果的情况下,其中第3组样品的电阻值为4.6×107 Ω∙cm2,第6组样品的电阻值为3.5×107 Ω∙cm2,因此,第3组样品适宜制备VCSEL器件。使用该工艺参数制备了VCSEL器件,得到的P-I-V曲线和近场分布如图9所示,该器件能实现较好的激光出射。
Study on proton implantation isolation of GaAs-based devices
-
摘要: 质子注入参数对注入型垂直腔面发射激光器(Vertical cavity surface emitting laser, VCSEL)的电流限制孔径位置及电流限制效果具有较大影响。文中从质子注入的能量和剂量及其相互作用对VCSEL电流限制孔径的影响规律及机制出发,通过理论模拟分析了注入参数对质子分布及注入区电阻值的影响。在此基础上,采用VCSEL外延片进行了质子注入实验研究。实验结果和理论分析均表明:注入区电流隔离效果及质子分布受注入能量和剂量共同调控。当注入参数为320 keV、8×1014 cm−2时,经430 ℃、30 s退火后可得到结深约0.7 μm,平均射程距有源区约1.3 μm,电阻值达4.6×107 Ω∙cm2的质子注入区。使用该参数制备的VCSEL器件实现了较好的激光激射,证明该质子分布不仅可避免VCSEL有源区损伤,而且能实现较好的电流隔离效果,满足VCSEL电流限制孔径的制备要求。研究结果对质子注入型VCSEL的芯片结构及工艺优化具有重要指导意义。Abstract: Proton injection parameters have a great influence on the position of current confinement aperture and the effect of current isolation of the implantation vertical cavity surface emitting laser (VCSEL). From the influence rule and mechanism of the energy and dose of proton implantation and their interaction on the current confinement aperture of VCSEL, this paper analyzes the influence of implantation parameters on the proton distribution and the resistance value of the implanted region by theoretical simulation firstly. And then proton implantation experimental research were carried out using VCSEL epitaxial wafers on this basis. Both the experimental results and theoretical analysis show that the current isolation effect and proton distribution in the injection region are controlled by injection energy and dose. When the implantation parameters are 320 keV and 8×1014 cm−2, after annealing at 430 ℃ for 30 s, a proton implantation region can be obtained with a junction depth of about 0.7 μm, an average range of about 1.3 μm from the active region and a resistance value of 4.6×107 Ω∙cm2. The VCSEL device was fabricated by using this parameter, and better laser excitation was achieved. It is proved that the proton distribution can not only avoid the damage of VCSEL active region, but also achieve a excellent current isolation effect, meeting the fabrication requirements of the VCSEL current confinement aperture. The results of this study have important guiding significance for the chip structure and process optimization of proton-injected VCSELs.
-
Key words:
- proton implantation /
- VCSEL /
- GaAs matrix /
- lattice defect
-
表 1 SEM 表征结果
Table 1. Characterization results of SEM
Group Energy/
keVDose/
cm−2Average range, $ {R_P} $ /μmJunction depth, $ {\text{2}}\Delta {R_P} $ /μmDistance from
active region,
d/μm1 280 6×1014 1.55 0.35 1.28 2 320 6×1014 1.67 0.61 1.03 3 320 8×1014 1.67 0.73 0.97 4 320 1×1015 1.68 0.84 0.9 5 320 2×1015 1.70 1.07 0.77 6 360 6×1014 1.72 0.71 0.93 7 400 2×1015 2.23 1.3 0 -
[1] Han T, Zhang Y, Miao C X, et al. Anti-interference temperature control for VCSEL laser [J]. Infrared and Laser Engineering, 2020, 49(9): 20190461. (in Chinese) [2] Elafandy R T, Kang J H, Mi C, et al. Study and application of birefringent nanoporous GaN in the polarization control of blue vertical-cavity surface-emitting lasers [J]. ACS Photonics, 2021, 8(4): 1041-1047. doi: 10.1021/acsphotonics.1c00211 [3] Wang H, Du G, Cui H, et al. Room temperature continuous wave low threshold current ion implanted vertical cavity surface emitting laser using tungsten wires as mask [J]. Optics & Laser Technology, 2003, 35(5): 341-344. [4] Tong H X, Tong C Z, Wang Z Y, et al. Advances in the technology of 850 nm high-speed vertical cavity surface emitting lasers (Invited) [J]. Infrared and Laser Engineering, 2020, 49(12): 20201077. (in Chinese) [5] Zhang Y Q, Zuo Z Y, Kan Q, et al. Common failure modes and mechanisms in oxide vertical cavity surface emitting lasers [J]. Chinese Optics, 2022, 15(2): 187-209. (in Chinese) [6] Pei L N, Zou Y G, Shi L L, et al. Design of low stress MEMS cantilever structure with tunable VCSEL [J]. Infrared and Laser Engineering, 2019, 48(4): 0420002. (in Chinese) [7] Liu C, Cao C F, Lao Y F, et al. Electrically confined aperture formed by ion implantation and its effect on device optoelectronic characteristic [J]. Journal of Semiconductors, 2008, 29(4): 765-769. (in Chinese) [8] Chen L, Luo Y, Feng Y, et al. Temperature dependence of wet oxidation process based on VCSEL [J]. Chinese Journal of Lasers, 2020, 47(7): 0701023. (in Chinese) [9] Naito H, Miyamoto M, Aoki Y, et al. Short-pulse operation of a high-power-density proton-implanted vertical-cavity surface-emitting laser array [J]. Applied Physics Express, 2012, 5(8): 082104. [10] Wang H S, Du G T, Cui H F, et al. Room temperature CW 850 nm vertical cavity surface emitting lasers fabricated by tilt ion implanting using Tungsten wires as mask [J]. Chinese Journal of Lasers, 2004(2): 129-132. (in Chinese) [11] Pan G Z. Study of large-scale in-phased vertical cavity surface emitting laser array and integrated phase-controlled beam steering chip[D]. Beijing: Beijing University of Technology, 2020. (in Chinese) [12] Mao M M, Xu C, Wei S M, et al. The effects of proton implant energy on threshold and output power of vertical cavity surface emitting laser [J]. Acta Physica Sinica, 2012, 61(21): 214207. (in Chinese) [13] Chen Z Q, Sun X W, Xia G Q, et al. Proton implant isolation of AlGaInP/GaAs heterojunction bipolar transistor structures in fabrication [J]. Acta Physica Sinica, 2000, 49(2): 375-378. (in Chinese) doi: 10.7498/aps.49.375 [14] Wang S W. Investigation of the radiation damage of a new PIN semi-conduct detector[D]. Lanzhou: Lanzhou University, 2006. (in Chinese) [15] Ge Z Y. The study on GaAs photoconductor detector by proton irradiation[D]. Chengdu: Sichuan University, 2004. (in Chinese) [16] Mao M M. Study of antiguide vertical cavity surface emitting laser and its two dimensional coherently coupled array[D]. Beijing: Beijing University of Technology, 2014. (in Chinese) [17] Que Z A, Hao X Y, He G K, et al. Simulation of ion implantation process for coaxial HpGe detector [J]. Nuclear Electronics & Detection Technology, 2020, 40(2): 239-243. (in Chinese)