The ionization of the diffuse ionized gas
The ionization of the diffuse ionized gas
复制标题
扩散电离气体的电离
DOI：
10.1086/174275
复制
发表时间：
1994
期刊：
影响因子：
--
通讯作者：
J. Mathis
中科院分区：
文献类型：
--
作者：
H. Domgorgen;J. Mathis
漫射电离气体（DIG）负责产生微弱但普遍存在的H（次α），其电离所需的能量是银河系所有O型恒星的15%。似乎没有其他可行的电离源。光谱在低电离阶段（N II， SII）强，在（O III）弱，在（O I） lambda 6300非常弱，至少在一个观测良好的方向上，都与密度较大的H II区有显著不同。我们给出了低密度、低激发的光电离模型来解释观测到的光谱。lambda 6300的观测消除了DIG （Mathis 1986）最简单的模型，在这个模型中，中性H延伸到非常稀释的恒星辐射场的边缘之外。我们目前的模型包括两个部分。其中一幅代表星际H I云的边缘，延伸到H变为中性的点。在第二种情况下，H（sup o）的分数不允许超过0.05 ~ 0.1。两者都具有非常低的电离参数值，或电离光子与电子的数量密度之比。我们的模型所要求的电离参数与漫射星际介质中光子和电子密度的观测值一致。星际尘埃在这两个模型中都不重要。该模型的预测是，银河系至少将本地O型恒星的电离辐射的4%泄漏到星系间介质中，即（O II） lambda 3727大约等于1.1 H（次α）， He(sup +)/H （sup +）大约等于0.6 He/H。这幅图的一个主要困难是电离辐射从银河系平面上的电离恒星传播到DIG。正如Miller & Cox（1993）所做的那样，我们认为，从O星的位置看，电离辐射在中性氢的分布中产生了空洞或隧道，而我们看到的天空被氢片或细丝覆盖，因为我们不在电离辐射源附近。在我们的模型中，在超过lambda 6300的观测极限之前，电离气体和中性气体之间的几个界面是允许的。另一个困难是高纬度恒星HD 93521的气相N(S(sup +))/N（H(sup o)）比（Spitzer & Fitzpatrick 1993）。我们要求（S/H）比太阳低一些，要么是因为谷物的消耗，要么是因为银河系的丰度梯度，这样S（sup +）就可以在每个单独的H（sup o）和H（sup +）区域产生。
The diffuse ionized gas (DIG), responsible for producing the faint but pervasive H(sub alpha), requires 15% of the power of all Galactic O stars for its ionization. No other source of ionization seems practical. The spectrum is strong in low stages of ionization (N II, SII), weak in (O III), and very weak in (O I) lambda 6300, at least in one well-observed direction, all significantly different from denser H II regions. We give low-density, low-excitation photoionization models that explain the observed spectrum. The lambda 6300 observation eliminates the simplest models for the DIG (Mathis 1986), in which neutral H extends beyond the edge of the very dilute stellar radiation field. Our present models include two components. One, representing the edges of interstellar H I clouds, extends to the point where H becomes neutral. In the second, the fraction of H(sup o) is not allowed to exceed 0.05 to 0.1. Both have very low values of the ionization parameter, or ratio of the number densities of ionizing photons to electrons. The ionization parameter required by our models is shown to be compatible with observed values of photon and electron densities in the diffuse interstellar medium. Interstellar dust is not important in either model. Predictions of the model are that the Galaxy is leaking about 4% of the ionizing radiation of at least the local O stars into the intergalactic medium, that (O II) lambda 3727 approximately equals 1.1 H(sub alpha), and that He(sup +)/H (sup +)) approximately equals 0.6 He/H. A major difficul ty of this picture is having ionizing radiation propagate from the ionizing stars in the plane of the Galaxy to the DIG. We suggest, as do Miller & Cox (1993), that the ionizing radiation from O stars produces holes or tunnels in the distribution of neutral H as seen from their positions, while we see a sky covered with H I sheets or filaments because we are not near a source of ionizing radiation. With our models, several interfaces between ionized and neutral gas are allowed before the observational limit on lambda 6300 is exceeded. Another difficulty is the gas-phase N(S(sup +))/N(H(sup o)) ratio toward the high-latitude star HD 93521 (Spitzer & Fitzpatrick 1993). We require (S/H) to be somewhat lower than solar, either because of depletion onto grains or a Galactic abundance gradient, so that the S(sup +) can be produced in each of separate H(sup o) and H(sup +) regions.