The Halo porphyry copper-molybdenum prospect is a porphyry system in Ratanakiri province, northeastern part of Cambodia. The province has a potential for porphyry-type deposits such as porphyry copper-molybdenum (Halo prospect, China Wall prospect) and porphyry copper-molybdenum-gold (Okalla prospect) (Figure 1(a); [1] ). The Halo porphyry copper-molybdenum prospect lies 2 km southeast of a strike slip fault trending NE-SW, known as the PhumSyarung-Dok Yong Fault corridor. The Halo prospect is hosted by intermediate to felsic intrusive and volcanic rocks. However, detailed geochemical characteristics of rocks and alteration lithogeochemistry were not studied yet. This is the first study that focuses on the characteristics of the intrusive rocks (major elements, trace elements, and rare earth elements) to constrain petrogenesis and tectonic setting, and lithogeochemistry of intrusive and volcanic rocks to demonstrate elements transportation during hydrothermal alteration at the Halo prospect. The main purpose of this study is to confirm the deposit type based on the datafrom by present study and to find the centre of porphyry deposit in the Halo prospect.

Cambodia is located in the southern part of Indochina Terrane, mainland of Southeast Asia. The mainland SE Asia comprises several Gondwana-derived terranes including Indochina, South China, Sibumasu and West Myanmar Terranes, which assembled and amalgamated by subduction-collision processes during the Late Palaeozoic to Mesozoic (Figure 1(a); [2] ). The Indochina Terrane is made up of several tectonic units, which host mineralization belts such as Truong Son Fold Belt (TSFB), Loei Fold Belt (LFB), Dalat-Kratie Belt (DKB) [3] . The Halo porphyry copper-molybdenum prospect in northeastern Cambodia lying on the DKB consists of Triassic to Cretaceous sedimentary rocks intruded by Cretaceous (125 - 75 Ma) volcano-plutonic rocks, which are overlain by Quaternary intraplate basalts. The DKB extends across the region from Cambodia to southern Vietnam and overlies the southern continuation of the TSFB and LFB

(Figure 1(a); [4] ). The Cretaceous belt is chronologically comparable to the plutono-volcanic rocks of the late stage of the Yanshanian Orogeny in SE China (140 - 65 Ma) [5] . Thus, this belt could have formed in a similar tectonic setting as a southern continuation of the Yanshanian belt, which was formed by the subduction of the Palaeo-Pacific Plate beneath the Eurasia Plate (including the Indochina Terrane) ( [5] & [6] ). The ore deposits in the DKB are intrusion-related gold (Figure 1(a); e.g. Okvau, Snoul in Cambodia [7] and TienThuan in southern Vietnam [8] ), porphyry copper-gold, skarn lead-zinc, sediment-hosted gold (e.g. North Kratie in Cambodia; [7] & [9] ), and porphyry copper-molybdenum, gold-silver epithermal veins over-printing porphyry copper-molybdenum-(gold) (Canada Wall and OkallaEast in Cambodia; [1] ) systems.

The Halo prospect is located on the edge of a monzogranite pluton to the south which is covered with a prominent silica cap outcrops on hilltops in the area. The host rocks of the Halo prospect are composed of felsic and intermediate intrusive and volcanic rocks. The volcanic rocks were intruded by diorite, granodiorite, and a quartz feldspar porphyry stock, which brought copper and molybdenite mineralization in veins and veinlets in both the intrusive and volcanic units [1] . The quartz feldspar porphyry exposed on the surface is medium to coarse-grained with abundant disseminated sulfides and appears to be the intrusion that brought in the molybdenum-copper mineralization. It is typically altered to silica, sericite, or sericite-chlorite [10] . Field observations of the core indicate that pyrite, molybdenite and chalcopyrite veinlets and veins exposed on surface exposures continue to the bottom of drill holes HD1 and HD2 (Figure 1(b)). Drill hole HD1 shows appreciable copper and silver mineralization in the top 99 metres, with anomalous levels of molybdenum corresponding to disseminated sulfides observed in the same interval. Included in the 99m interval is an intercept of 2345 ppm Cu, 1.34 g/t Ag, as well as 261.4 ppm Mo over 88.9 m which includes 7.9 m of 8043 ppm Cu, 2.24 g/t Ag, and 320.7 ppm Mo from 10.1 to 18.0 m [11] .

A major magnetic, concentrically ringed “doughnut” anomaly is present in the central part of the Halo porphyry copper-molybdenum prospect [10] . This magnetic anomaly is a typical expression of a porphyry system with a magnetic high in the center surrounded by a zone of lower magnetic response. The high magnetism typically represents an increase of the abundance of magnetite associated with the mineralization in the centre of porphyry type deposit, with surrounding low magnetism which reflects a depletion of the abundance of magnetite in the mineralized host rocks [12] . Termite mound geochemistry revealed anomalies of copper and molybdenum at the central part of Halo (Figure 1(b)) with the high up to 700 ppm Cu while the anomaly is generally >250 ppm Cu, covering 1.5 km extending WNW-ESE; 1 km extending NNE-SSW. High anomaly of molybdenum is 400 ppm in 1.2 km by 75 m, extending EW-NS from central part of Halo. Rock chip geochemistry ranges up to 7470 ppm Cu up to 1450 ppm Mo at the central part of Halo (Figure 1(b); [1] ). The main alteration zones developed in this prospect on the surface are inner phyllic (sericite), outer phyllic (sericite-chlorite), argillic (kaolinite-dickite-pyrophyllite-halloysite), propylitic (chlorite-epidote-quartz-calcite) and silicic alteration (quartz) according to field observation and short wave infrared spectroscopy on alteration mineral identification (Figure 2(a); [10] ). Phyllic alteration is located the central part of the Halo prospect and was overprinted with silicic alteration (outcrop sample on the hilltops, up to 15 m) [10] .

Whole-rock major elements oxides, trace and rare earth elements concentrations of 42 samples including diorite, granodiorite, quartz feldspar porphyry, hornblende diorite collected from surface and four drill holes HD1, HD2, HD3 and HD4 (Figure 2(b)) were analyzed by X-ray fluorescence (XRF) spectroscopy using a RIGAKU RIX-3100 and by inductively coupled mass spectrometry (ICP-MS) Alilent Technologies 7500, respectively, at the Center of Advanced Instrumental Analysis, Kyushu University. Loss on ignition (LOI) was determined by heating the samples at 1000˚C for 2 hours to determine relative weight loss. Thirty petrographic thin sections were prepared to identify textural characteristics and alteration minerals of rocks and observed using a Nikon Eclipse E600

POL microscope equipped with an AdvanCam-U3II camera. Both bulk and clay fraction X-ray diffraction analysis were conducted to identifyalteration mineral assemblage using a RigakuUltima IV X-ray diffractometer at Department of Earth Resources Engineering, Kyushu University.

In order to determine textural and mineral composition of each rock type, petrography was conducted on the relatively fresh samples, least altered samples and also altered samples including outcrop samples and drill core samples.

Whole-rock compositions of 35 intrusive rocks and 7 volcanic rocks from the Halo prospect (Location of sample in Figure 1 and Figure 2) are determined in Table 1. The effects of hydrothermal alteration on whole-rock geochemical compositions have been assessed using a combination of petrography and bivariate plots (Figure 4(a)). Thus, samples from the Halo prospect were plotted on an alteration box plot diagram using the alteration index (AI) against the chlorite-carbonate-pyrite index (CCPI) proposed by [13] . The alteration index (AI) was calculated by 100 times of the ratio of the sum of K₂O and MgO by the sum of K₂O, MgO, Na₂O, and CaO. The chlorite-carbonate-pyrite index (CCPI) were calculated by 100 times of the ratio of the sum of MgO and FeO by the sum of MgO, FeO, Na₂O, and K₂O. All major elements used to calculate AI and CCPI are expressed as wt.%. The samples plotted on the alteration box plot diagram display 1) moderate K-feldspar-sericite, 2) sericite, 3) sericite-chlorite-pyrite, and 4) weak epidote-calcite ± albite alteration trends. However, the majority of the samples do not significantly deviate from the least-altered box (Figure 4(a)). LOI (loss on ignition) values of least altered rocks range from 1.5% to 5.6%, while LOI values of the altered rock range from 1.3% to 4.8%. Thus the LOI values may misclassify between least altered and altered sample in this study. However, the result of least altered rocks classified by box plot [13] is consistent with the petrography. The SiO₂ concentrations of intrusive rocks and volcanic rocks in the Halo prospect range from 57.7 to 75.6 wt.% and 61.7 to 68.2 wt.%, respectively. The (Na₂O+K₂O) versus SiO₂ diagram [14] indicates that the intrusive rocks in the Halo are classified as diorites, granodiorites and granites (quartz feldspar porphyry) (Figure (4b)). The high field strength element (HFSE) bivariate plot of Zr/TiO2–Nb/Y [15] is used to distinguish geochemically the various volcanic rocks from the Halo prospect (Figure 5(a)). The volcanic rocks plot in rhyodacite/dacite to trachyandesite fields. The rocks classification are reliable for diorite and granodiorite as they are the least altered. The quartz feldspar porphyry spatially associated with mineralization, show strong alteration in box

plots. Therefore, major elements are deemed unreliable to classify the rock types as strongly altered.

On the SiO₂ versus K₂O discrimination diagram for intrusive and volcanic rocks (Figure 5(b)), intrusive rocks are medium-K calc-alkaline, high-K calc-alkaline and shoshonitic fields [16] . The samples with high-K values contain hydrothermal K-feldspar and biotite, implying that K was added to the rocks during hydrothermal alteration (potassic alteration). Thus, it is interpreted that the high-K trend is a result of potassic alteration and does not reflect primary igneous composition (e.g. shoshonite). The least altered granodiorite (GRDP2) plots in high-K calc-alkaline to shoshonite. Therefore, the primary igneous compositions of rock belong to the high-K calc-alkaline series. The least altered diorite and granodiorite exposed on surface plot in medium-K to high-K calc-alkaline field. Thus the primary composition of diorite and granodiorite belongs to medium-K calc-alkaline series. Diorite and granodiorite (DI3 and GRD3) plot on boundary between medium-K and high-K calc-alkaline field. Least altered hornblende diorite (HDI4) plots in medium-K calc-alkaline field except one sample with strong alteration plot in low-K calc-alkaline filed. The primary igneous compositions of the hornblende diorite belong to the high-K calc-alkaline series. In contrast, low-K composition is also affected by hydrothermal alteration. The granodiorite (HGRD4) plots on boundary between medium-K to high-K calc-alkaline, is close to primary igneous rock compositions. Based on the box plot and SiO₂ versus K₂O (wt.%) diagram, some samples are strongly affected by K mobility. Therefore, we further use Pearce element ratio (PER) analysis to discuss in next section in order to elucidate about alteration lithogeochemistry [17] .

Chondrite-normalized REE patterns [18] of the intrusive and volcanic rock from the Halo prospect (diorite, granodiorite, granodiorite porphyry, quartz feldspar porphyry, dacite and andesite) show general enrichment of light REEs (LREE) ((La/Yb)N) = 4.4 to 39.3) and depletion of heavy REEs (HREE) ((Gd/Yb)N = 0.79 to 3.59) (Figure 6). Distinct weakly negative Eu anomaly (Eu/*Eu = 0.65 to 0.98) of diorite, granodiorite, and granodiorite porphyry would be indicative of plagioclase fractionation [19] . Quartz feldspar porphyry, dacite and andesite porphyry characterized by negative to positive Eu anomalies (Eu/*Eu = 1.15 to 2.41). Negative of Eu anomalies indicates possible fractionation of Eu to plagioclase, while the positive Eu anomalies suggest that Eu present in primary plagioclase are immobile during potassic alteration, while other REE were slightly removed during potassic alteration. Thus, REE patterns are slightly mobile during alteration. The primitive mantle-normalized trace elements diagram of intrusive and volcanic rocks in Halo are characterized by enrichments of large-ion lithophile elements (LILE, such as K, Rb and Ba) and depletions of high field strength elements (HFSE, such as Nb and Ti) (Figure 6).

Quartz feldspar porphyry samples (QFP1-1 to QFP1-4) contain higher total REE concentration ranging from 44.8 to 54.6 ppm compared to quartz feldspar porphyry samples (QFP1-5 to QFP1-10) which contain lower total REE concentration ranging from 16.4 to 35.9ppm. Based on petrography, all quartz feldspar porphyries are potassicaly altered and samples QFP1-1 to QFP1-4 are less altered than the samples QFP1-5 to QFP1-10. This suggests that the REE was slightly removed during potassic alteration, except for Eu which was accommodated in plagioclase and immobile during potassic alteration.

General element ratios (GERs) and Pearce element ratios (PERs) are both molar element ratios (MERs) used to depict geochemical processes such as alteration ( [17] & [20] ). PER analysis uses molar element concentrations ratio to the molar concentration of an element which has remained unchanged during mass transfer processes, i.e. a “conserved” element ( [21] & [22] ). Thus in order to quantify the intensity of hydrothermal alteration around major mineralized zones, it is necessary to use a technique that distinguish the metasomatic impact of the mineralizing fluids, a prerequisite for which is pre-existing geochemical heterogeneity in the host rocks. It is also necessary to discriminate between the effects of hydrothermal activity and unrelated events such as weathering or metamorphism. A conserved element must be identified for use as a denominator in the PERs. Conserved elements are also used to test the cogenetic character of the rocks and group them accordingly. A PER diagram of Pr versus Nd shows the same behavior during alteration and a best-fit line with a positive slop that intersects the origin indicating that the rocks are derived from a single precursor (Figure 7(a); [23] ). The samples are cogenetic, thus PER can be adapted [22] .

A PER plot of (2Ca + Na + K) versus Al/Nd (Figure 7(b)) discriminates between unaltered and hydrothermally altered felsic rocks. The least altered samples plot closer to the feldspar-plagioclase control line with slop of 1 from the origin while the completely sericitized samples plot closer to the line with slop of 1/3 from the origin [24] . The dacite samples plot around and toward the muscovite control line with slop 1/3, suggesting the dacite has been affected by quartz-sericite-pyrite (phyllic) alteration (Figure 7(b)) while the quartz feldspar porphyry, andesite porphyry, diorite and granodiorite samples plot between the plagioclase and the muscovite control lines, suggesting that these rocks have been affected by varying degree of sericite alteration. This is consistent with plagioclase replaced by sericite observed in petrography. Potassic alteration such as secondary biotite or secondary K-feldspar in rocks is not discriminated from unaltered rocks on this diagram. Epidote alteration is represented by the control line with slope 16/3. The hornblende diorite samples plot between the epidote and plagioclase control lines consistent with petrography and XRD analysis.

1) The Halo porphyry copper molybdenum prospect is hosted in diorite, granodiorite, quartz feldspar porphyry, granodiorite porphyry, andesite porphyry, mafic dykes, aplite dykes, and felsic dykes. Quartz feldspar porphyry and granodiorite porphyry are considered as the main intrusive body and most favorable host copper-molybdenum mineralization. They intersected drill hole HD1 and HD2, which are associated with potassic alteration and dominated by Cu-Fe sulfide mineralization, quartz vein stockworks and magnetite veins.

2) The diorite and granodiorite belong to medium-K calc-alkaline to high-K calc-alkaline series while rock-series of volcanic rocks and quartz feldspar porphyries were not determined due to the mobility of K during alteration. The intrusive rocks in the Halo prospect display geochemical features of trace elements typical of magmatism in a subduction zone, such as enrichment of LILE and negative anomalies of Nb and Ti. The intrusive rocks and volcanic rocks plot within the domains of volcanic arc granite based on spider diagram, (Y + Nb) versus Rb diagram, and Y versus Nb diagram. Lithochemistry of intrusive and volcanic rocks in the Halo porphyry copper-molybdenum prospect is consistent with the rocks associated with porphyry deposits type ore in the world generated by magmatism in subduction setting related.

3) General element ratios and Pearce element ratios (PER) define three alterations in the Halo prospect such as potassic (secondary K-feldspar and biotite), phyllic (sericite) and propylitic (epidote) alteration. The alteration index of the dacite and the quartz feldspar porphyry calculated from PER are higher than that of the granodiorite porphyry. This is consistent with the high copper concentration of the dacite and the quartz feldspar porphyry up to 7048 ppm. The highly altered zone vectoring center of the hydrothermal system may represent the locus of mineralization.

4) According to result mentioned above, the Halo prospect is a porphyry copper-molybdenum prospect and drill hole HD1 and HD2 are situated close to the centre of porphyry deposit than drill hole HD3 and HD4.

The authors wish to express our special thanks to Professor Akira Imai, Department of Earth Resources Engineering, Kyushu University in Japan for his kind support and encouragement to carry out this research work, advice and reading the manuscript and give facilities in varies ways. We would like to express our deepest thanks to Mr. John Dau Pau, a Vice President of Operation, Mr. Mike Weeks, a chairman in Angkor Gold, for allowing and supporting to the field research in the Halo prospect, Oyadao South area, Ratanakiri, Cambodia. We would like to thanks also for Craig T. Richardson for is excellent master thesis study made on PhumSyarung-Dok Yong Fault Corridor (Cambodia), and the exceptional cooperation of Dennis Ouellet (Exploration Manager of Angkor Gold Corp.). We would like to express our gratitude to the ASEAN University Network/Southeast Asia Engineering Education Development Network (AUN/SEED-NET), Japanese International Cooperation Agency (JICA) for the PhD scholarship. We would like to special thank JSPS KAKENHI Grants Number 18H01927 and JSPS Core-to-CoreProgram provide us financial support during field work. I am also indebted to all lecturers and Dr. NallisKry, a head of the Faculty of Geo-resources and Geotechnical Engineering, Institute of Technology of Cambodia, Cambodia, who has given me the constant support and guidance for my PhD at Kyushu University.

The authors declare no conflicts of interest regarding the publication of this paper.

Sirisokha, S., Yonezu, K., Tindell, T., Nakanishi, T., Watanabe, K. and Pelletier, J. (2019) Lithogeochemistry of Intrusive Rocks in the Halo Porphyry Copper-Molybdenum Prospect, Northeast Cambodia. Open Journal of Geology, 9, 342-363. https://doi.org/10.4236/ojg.2019.97023