1: \begin{abstract}

2:  The complex Gaussian Multiplicative Chaos (or complex GMC) is informally defined as  a random measure $e^{\gamma X} \dd x$ where $X$ is a log correlated Gaussian field on $\bbR^d$  and $\gamma=\alpha+i\beta$ is a complex parameter. The correlation function of $X$ is of the form 

3:  $$ K(x,y)= \log \frac{1}{|x-y|}+ L(x,y),$$

4: where  $L$ is a continuous function. In the present paper, we consider the cases $\gamma\in \cP_{\mathrm{I/II}}$ and $\gamma\in  \cP'_{\mathrm{II/III}}$

5:  where 

6:  $$  \cP_{\mathrm{I/II}}:= \{ \alpha+i \beta \ : \alpha,\beta \in \bbR \ ; |\alpha|>|\beta| \ ; \ |\alpha|+|\beta|=\sqrt{2d}  \},

7:  $$ 

8:  and 

9:   $$  \cP'_{\mathrm{II/III}}:= \{ \alpha+i \beta \ : \alpha,\beta \in \bbR \ ; \ |\alpha|= \sqrt{d/2} \ ; \ |\beta|>\sqrt{2d}  \},

10:  $$

11:  We prove that if $X$ is replaced by an approximation $X_\gep$ obtained via mollification, then  $e^{\gamma X_\gep} \dd x$, when properly rescaled, converges when $\gep\to 0$. The limit does not depend on the mollification  kernel. When $\gamma\in \cP_{\mathrm{I/II}}$, the convergence holds in probability and in $L^p$ for some value of $p\in [1,\sqrt{2d}/\alpha)$. When $\gamma\in \cP'_{\mathrm{II/III}}$

12:  the convergence holds only in law. In this latter case, the  limit can be described

13:  a complex Gaussian white noise with a random intensity given by a critical real GMC. The regions $\cP_{\mathrm{I/II}}$ and $\cP'_{\mathrm{II/III}}$

14:  correspond to phase boundary between the three different regions of the complex GMC phase diagram. These results complete previous results obtained for the GMC in phase I \cite{lacoin2020} and III \cite{MR4413209} and only leave as an open problem the question of convergence in phase II.

15:    \\[10pt]

16:   2010 \textit{Mathematics Subject Classification: 60F99,  	60G15,  	82B99.}\\

17:   \textit{Keywords: Random distributions, $\log$-correlated fields, Gaussian Multiplicative Chaos.}

18:  \end{abstract}

19: